INDEX

Explanations

words related to differentiating or comparing items

oai_token-act-pair · gpt-3.5-turbo

terms related to accountability or behavior standards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

quickShipAvailable

-0.81

dit

-0.76

dates

-0.74

Buzz

-0.73

JPM

-0.72

bler

-0.71

 Leaves

-0.71

forth

-0.70

 Relief

-0.70

Oo

-0.70

POSITIVE LOGITS

èĢħ

0.78

auga

0.74

 roaming

0.72

 conduc

0.70

 conduct

0.64

 indemn

0.64

 discontin

0.64

 rods

0.63

 polymorph

0.62

 leasing

0.62

Activations Density 0.000%

words related to differentiating or comparing items

terms related to accountability or behavior standards

No Comments

No Known Activations