INDEX

Explanations

expressions related to agreement or unanimity of opinions

oai_token-act-pair · gpt-3.5-turbo

references to general agreement or shared viewpoints

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

hib

-0.85

gin

-0.80

chester

-0.77

strip

-0.77

ikarp

-0.76

thy

-0.73

vern

-0.71

undai

-0.70

duc

-0.70

gun

-0.70

POSITIVE LOGITS

 consensus

1.02

 unanimously

0.83

ensus

0.78

 amongst

0.76

IFIED

0.75

itarian

0.74

 opinion

0.72

 among

0.71

ãĥ¼ãĥĨ

0.71

Activations Density 0.026%

expressions related to agreement or unanimity of opinions

references to general agreement or shared viewpoints

No Comments

No Known Activations