INDEX

Explanations

adjectives related to improvement or intensification

oai_token-act-pair · gpt-3.5-turbo

terms associated with improvement or enhancement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ships

-0.81

atre

-0.73

PLIED

-0.72

ices

-0.72

tips

-0.72

wered

-0.71

oos

-0.71

emi

-0.71

Scale

-0.69

VPN

-0.69

POSITIVE LOGITS

 versions

0.95

 version

0.84

 opportunity

0.76

 scrutiny

0.76

 depreciation

0.74

 embodiment

0.74

 edition

0.74

 senses

0.73

 immunity

0.73

 efficiency

0.73

Activations Density 0.094%

adjectives related to improvement or intensification

terms associated with improvement or enhancement

No Comments

No Known Activations