INDEX

Explanations

numerical values expressed as percentages

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ailability

-0.72

 conduc

-0.69

 ende

-0.66

 Flavoring

-0.64

 traged

-0.64

aukee

-0.61

uctor

-0.61

 culmin

-0.60

 enthusi

-0.60

pmwiki

-0.59

POSITIVE LOGITS

Â±

0.68

ãĥł

0.61

font

0.60

×Ĳ

0.58

olla

0.57

=~=~

0.55

×Ļ

0.55

aunt

0.54

 Downloads

0.54

Ö

0.54

Activations Density 0.155%

numerical values expressed as percentages

numerical data or statistics

No Comments

No Known Activations