INDEX

Explanations

phrases related to certain situations or scenarios

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate legal or formal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

uristic

-0.78

uliffe

-0.77

venants

-0.74

earchers

-0.74

ient

-0.73

Cause

-0.72

arak

-0.72

igue

-0.71

assian

-0.70

arna

-0.70

POSITIVE LOGITS

resp

0.64

 Sterling

0.61

 instance

0.60

 cases

0.60

 ours

0.59

 needing

0.58

gotten

0.58

gery

0.57

Zak

0.57

Ms

0.57

Activations Density 0.105%

phrases related to certain situations or scenarios

phrases that indicate legal or formal contexts

No Comments

No Known Activations