INDEX

Explanations

numerical fractions or percentages

oai_token-act-pair · gpt-3.5-turbo

quantitative phrases indicating frequencies or probabilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

complex

-0.79

sort

-0.71

imir

-0.71

ãĤ´ãĥ³

-0.67

gang

-0.65

ande

-0.65

unity

-0.65

rev

-0.65

unct

-0.65

Stage

-0.64

POSITIVE LOGITS

 Americans

0.97

 respondents

0.96

 households

0.88

 earners

0.85

 adults

0.81

 cases

0.79

 Australians

0.78

 Britons

0.78

 chance

0.75

 adult

0.73

Activations Density 0.052%

numerical fractions or percentages

quantitative phrases indicating frequencies or probabilities

No Comments

No Known Activations