INDEX

Explanations

phrases expressing affirmation or agreement

oai_token-act-pair · gpt-3.5-turbo

negative contractions indicating denial or negation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Skydragon

-0.80

 Samar

-0.75

Kaf

-0.61

 Gmail

-0.61

ZIP

-0.61

 Hawaii

-0.60

MUS

-0.59

 PHOTO

-0.58

 Honolulu

-0.57

 Android

-0.57

POSITIVE LOGITS

thy

0.85

else

0.82

sure

0.81

ski

0.81

tu

0.78

cum

0.78

0.77

bec

0.77

shall

0.77

felt

0.75

Activations Density 0.253%

phrases expressing affirmation or agreement

negative contractions indicating denial or negation

No Comments

No Known Activations