INDEX

Explanations

phrases related to agreement or disagreement

oai_token-act-pair · gpt-3.5-turbo

phrases indicating disagreement or criticism

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

ario

-0.68

atto

-0.65

iture

-0.65

 notwithstanding

-0.62

 Donation

-0.61

etermination

-0.60

 solution

-0.60

leep

-0.60

 disposed

-0.59

urated

-0.59

POSITIVE LOGITS

 those

0.72

those

0.69

 these

0.68

our

0.68

us

0.64

them

0.64

their

0.64

Adams

0.61

 their

0.60

his

0.60

Activations Density 0.150%

phrases related to agreement or disagreement

phrases indicating disagreement or criticism

No Comments

No Known Activations