INDEX

Explanations

contractions of specific verbs and pronouns

oai_token-act-pair · gpt-3.5-turbo

negations or words indicating inability or denial

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

 eleph

-0.92

anwhile

-0.87

 pione

-0.85

 newcom

-0.80

senal

-0.79

ccording

-0.75

aditional

-0.74

Þ

-0.74

 conclud

-0.73

 enthusi

-0.73

POSITIVE LOGITS

't

1.58

ned

0.91

ÃŃ

0.90

Õ

0.90

Â´

0.78

eness

0.78

uts

0.76

nen

0.74

ny

0.74

ovan

0.74

Activations Density 0.155%

contractions of specific verbs and pronouns

negations or words indicating inability or denial

No Comments

No Known Activations