INDEX

Explanations

mentions of headwear items, particularly hats

oai_token-act-pair · gpt-3.5-turbo

the word "hat" and its variations in different contexts

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

hyde

-0.75

 Vaugh

-0.65

ULTS

-0.64

lect

-0.64

FORMATION

-0.62

idge

-0.62

naire

-0.62

taining

-0.62

á½

-0.62

ADE

-0.61

POSITIVE LOGITS

chet

1.07

chery

1.04

ia

0.87

ullah

0.80

tha

0.77

ibu

0.77

ched

0.76

ewater

0.76

ulhu

0.76

nikov

0.74

Activations Density 0.013%

mentions of headwear items, particularly hats

the word "hat" and its variations in different contexts

No Comments

No Known Activations