INDEX

Explanations

proper nouns or names with specific attributes, as evidenced by high activations near certain sequences of characters like "Sz" and "Sagan"

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those with the last name "Sz" or "Sagan."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

naissance

-0.83

tops

-0.65

phrine

-0.63

 steering

-0.63

eering

-0.63

acting

-0.62

 doub

-0.62

viously

-0.62

heads

-0.61

 Hurricanes

-0.61

POSITIVE LOGITS

Sz

1.22

ittal

0.87

ÃįÃį

0.87

Gy

0.81

Pi

0.79

iky

0.78

cz

0.77

Hung

0.73

pling

0.73

achu

0.73

Activations Density 0.015%

proper nouns or names with specific attributes, as evidenced by high activations near certain sequences of characters like "Sz" and "Sagan"

references to specific individuals, particularly those with the last name "Sz" or "Sagan."

No Comments

No Known Activations