INDEX

Explanations

the word "only" followed by a numerical value

oai_token-act-pair · gpt-3.5-turbo

instances of the word "a" appearing 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

Edit

-0.67

uploads

-0.66

lees

-0.65

cv

-0.65

Catal

-0.64

HMS

-0.63

ultimate

-0.61

sylvania

-0.61

ographs

-0.60

achine

-0.59

POSITIVE LOGITS

 handful

1.34

few

1.19

 fraction

1.18

 couple

1.02

 dozen

1.00

tad

0.95

 glimpse

0.94

 snippet

0.90

 hint

0.89

few

0.88

Activations Density 0.137%

the word "only" followed by a numerical value

instances of the word "a" appearing in various contexts

No Comments

No Known Activations