INDEX

Explanations

phrases related to actions and events, especially in a narrative context

oai_token-act-pair · gpt-3.5-turbo

sequences that indicate queries or requests for information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

erenn

-0.77

oppable

-0.72

enegger

-0.65

cffff

-0.63

 seiz

-0.63

 Seym

-0.62

 coerc

-0.61

bledon

-0.60

avorite

-0.60

nat

-0.59

POSITIVE LOGITS

isations

0.60

malink

0.57

CLOSE

0.57

lishes

0.56

ims

0.54

join

0.54

 thumbnail

0.53

>>

0.53

OULD

0.53

 nutshell

0.52

Activations Density 0.305%

phrases related to actions and events, especially in a narrative context

sequences that indicate queries or requests for information

No Comments

No Known Activations