INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

references to time and the actions that occur within specific time frames

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

onto

-0.77

lez

-0.73

alls

-0.73

ict

-0.68

irms

-0.68

 drown

-0.68

osite

-0.68

uala

-0.67

xit

-0.65

igr

-0.65

POSITIVE LOGITS

 regarded

0.69

 technically

0.67

 considered

0.66

 fashionable

0.66

 believed

0.66

 hadn

0.65

 nascent

0.64

 unknown

0.64

 Medline

0.63

 researching

0.62

Activations Density 0.183%

proper nouns

references to time and the actions that occur within specific time frames

No Comments

No Known Activations