INDEX

Explanations

images or visuals described in text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the" in various contexts throughout the text

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

nesty

-0.81

uthor

-0.76

arians

-0.74

LEASE

-0.73

HCR

-0.72

mart

-0.71

olicy

-0.70

thood

-0.70

hun

-0.70

cially

-0.69

POSITIVE LOGITS

 ceiling

1.04

 smallest

1.04

 vicinity

1.01

 slightest

1.00

 aforementioned

0.98

 periphery

0.98

 edges

0.97

 same

0.97

 entirety

0.96

 nearest

0.96

Activations Density 0.572%

images or visuals described in text

instances of the word "the" in various contexts throughout the text

No Comments

No Known Activations