INDEX

Explanations

the word "find" or variations of it

oai_token-act-pair · gpt-3.5-turbo

expressions of discovery or realization

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

idium

-0.85

 concess

-0.69

cussion

-0.60

 partic

-0.59

isoft

-0.59

ivari

-0.58

 Mandatory

-0.58

haw

-0.56

primary

-0.56

 coerc

-0.56

POSITIVE LOGITS

ãĤ¤ãĥĪ

0.84

ä½¿

0.77

ById

0.75

 fault

0.71

 Aires

0.71

ibility

0.71

 fertile

0.71

omething

0.69

out

0.68

½

0.68

Activations Density 0.078%

the word "find" or variations of it

expressions of discovery or realization

No Comments

No Known Activations