INDEX

Explanations

words related to expressing thoughts or opinions

oai_token-act-pair · gpt-3.5-turbo

references to personal thoughts and opinions

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

ALL

-0.78

 Mamm

-0.74

toe

-0.65

ARDS

-0.65

 Breach

-0.63

Naz

-0.61

 Adin

-0.59

 Peninsula

-0.58

two

-0.58

 Annex

-0.57

POSITIVE LOGITS

fulness

0.97

ileaks

0.92

 aloud

0.85

cience

0.83

 thoughts

0.83

ets

0.80

mith

0.78

reprene

0.77

 provoking

0.77

ynthesis

0.77

Activations Density 0.022%

words related to expressing thoughts or opinions

references to personal thoughts and opinions

No Comments

No Known Activations