INDEX

Explanations

quantitative information relating to things being reduced or decreased

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "fewer."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Draft

-0.75

MO

-0.73

Mut

-0.72

MAG

-0.69

SA

-0.68

 Editor

-0.67

Lo

-0.65

agency

-0.65

ESE

-0.65

MD

-0.64

POSITIVE LOGITS

 than

1.29

than

0.85

 Than

0.83

 calories

0.78

 digits

0.75

 "$:/

0.75

cible

0.73

 unwanted

0.70

 bytes

0.70

 manageable

0.70

Activations Density 0.013%

quantitative information relating to things being reduced or decreased

occurrences of the word "fewer."

No Comments

No Known Activations