INDEX

Explanations

the word "reason" indicating justification or cause

oai_token-act-pair · gpt-3.5-turbo

phrases that assert the necessity or absence of justification

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

 Carbuncle

-0.76

 Riders

-0.70

ibaba

-0.68

 Ping

-0.67

yss

-0.66

 puck

-0.65

Lag

-0.64

 Sour

-0.64

Lup

-0.63

 Roller

-0.63

POSITIVE LOGITS

why

1.01

abl

0.96

Reviewer

0.88

¿½

0.87

WHY

0.86

forward

0.85

 justifying

0.85

why

0.82

reason

0.81

="#

0.77

Activations Density 0.017%

the word "reason" indicating justification or cause

phrases that assert the necessity or absence of justification

No Comments

No Known Activations