INDEX

Explanations

phrases indicating exclusivity or limitation

oai_token-act-pair · gpt-3.5-turbo

phrases indicating exclusivity or restriction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

insula

-0.64

idon

-0.64

 Malf

-0.62

arted

-0.61

cent

-0.61

Nev

-0.59

 Massive

-0.59

put

-0.57

--------------------------------------------------------

-0.57

 charism

-0.57

POSITIVE LOGITS

 marginally

0.98

ICES

0.81

ices

0.80

 kidding

0.78

 incidentally

0.77

 spor

0.72

isons

0.69

 briefly

0.67

 mildly

0.67

 subset

0.66

Activations Density 0.056%

phrases indicating exclusivity or limitation

phrases indicating exclusivity or restriction

No Comments

No Known Activations