INDEX

Explanations

numeric values with decimal points

oai_token-act-pair · gpt-3.5-turbo

articles and the usage of the word "a" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

evidence

-0.66

 Jagu

-0.62

 Vaugh

-0.62

grounds

-0.57

eyes

-0.56

iments

-0.56

Eag

-0.56

ographs

-0.56

acebook

-0.56

Edit

-0.56

POSITIVE LOGITS

lot

1.00

 bunch

0.90

 couple

0.82

 handful

0.78

few

0.76

 huge

0.76

 glimpse

0.75

uras

0.73

 whopping

0.73

ird

0.72

Activations Density 0.558%

numeric values with decimal points

articles and the usage of the word "a" in various contexts

No Comments

No Known Activations