INDEX

Explanations

phrases related to communication or conversation

oai_token-act-pair · gpt-3.5-turbo

negative contractions, particularly the word "don't"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ãĥ¯ãĥ³

-0.74

çĶ°

-0.73

Buyable

-0.71

é¾

-0.70

 Palest

-0.70

è£ħ

-0.69

OSP

-0.64

 panc

-0.63

Interstitial

-0.62

Desk

-0.62

POSITIVE LOGITS

£

0.79

âĢķ

0.76

ump

0.75

Â¯

0.75

said

0.74

º

0.71

ares

0.71

you

0.70

¼

0.70

ooks

0.69

Activations Density 0.336%

phrases related to communication or conversation

negative contractions, particularly the word "don't"

No Comments

No Known Activations