INDEX

Explanations

phrases related to legal or procedural language

oai_token-act-pair · gpt-3.5-turbo

phrases related to medical conditions and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.69

 shepherd

-0.62

Nau

-0.62

ews

-0.60

 fridge

-0.60

 Saras

-0.59

riel

-0.58

luaj

-0.57

 passer

-0.57

 Nasa

-0.56

POSITIVE LOGITS

 secondly

1.16

 furthermore

0.97

 alternatively

0.93

 moreover

0.86

Therefore

0.82

 consequently

0.78

'(

0.75

``(

0.73

 preferably

0.72

particularly

0.71

Activations Density 0.121%

phrases related to legal or procedural language

phrases related to medical conditions and their consequences

No Comments

No Known Activations