INDEX

Explanations

words related to physical objects or locations

oai_token-act-pair · gpt-3.5-turbo

nouns related to entities or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ECA

-0.74

:{

-0.74

imon

-0.71

},{"

-0.71

innie

-0.70

>>>>

-0.67

EngineDebug

-0.67

%]

-0.66

pite

-0.66

è¦

-0.66

POSITIVE LOGITS

's

1.27

 itself

0.97

ÃŃs

0.85

wright

0.76

wide

0.70

 progressively

0.68

scape

0.65

 reeling

0.64

eers

0.61

runners

0.61

Activations Density 0.222%

words related to physical objects or locations

nouns related to entities or locations

No Comments

No Known Activations