INDEX

Explanations

specific characters or patterns within the text

oai_token-act-pair · gpt-3.5-turbo

general statements or claims

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

aeda

-0.64

 favoring

-0.59

Â¶

-0.59

baugh

-0.58

 utilizing

-0.58

 favorably

-0.58

wolves

-0.55

 respectively

-0.53

 favor

-0.53

 bragging

-0.52

POSITIVE LOGITS

 Inqu

0.75

Labour

0.71

 Minister

0.66

MLA

0.66

 councillor

0.66

gex

0.61

 Pearce

0.61

Coun

0.60

CPC

0.60

 Labour

0.59

Activations Density 0.255%

specific characters or patterns within the text

general statements or claims

No Comments

No Known Activations