INDEX

Explanations

mentions of specific locations or activities related to physical work and efficiency

oai_token-act-pair · gpt-3.5-turbo

instances of the word "in."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

spell

-0.69

Zar

-0.69

misc

-0.66

rim

-0.62

 Merry

-0.62

 Slime

-0.60

 grammar

-0.59

capital

-0.59

version

-0.58

cale

-0.56

POSITIVE LOGITS

ewitness

0.79

anche

0.73

pees

0.72

idav

0.70

Ò

0.67

ritional

0.67

chwitz

0.66

 exposures

0.66

ESCO

0.66

assetsadobe

0.66

Activations Density 0.000%

mentions of specific locations or activities related to physical work and efficiency

instances of the word "in."

No Comments

No Known Activations