INDEX

Explanations

information or items listed as the first step or initial point

oai_token-act-pair · gpt-3.5-turbo

instances of the word "starters."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

asca

-0.78

erry

-0.73

oÄŁ

-0.71

ilated

-0.71

orate

-0.71

ipl

-0.69

redit

-0.68

 Neurolog

-0.68

orie

-0.68

fax

-0.67

POSITIVE LOGITS

 starters

0.97

starter

0.87

starting

0.85

 starter

0.79

 pitchers

0.79

 offender

0.76

\<

0.73

 thereof

0.71

éļ

0.70

pring

0.69

Activations Density 0.003%

information or items listed as the first step or initial point

instances of the word "starters."

No Comments

No Known Activations