INDEX

Explanations

phrases related to specific actions or descriptions of events occurring

oai_token-act-pair · gpt-3.5-turbo

the presence of an end-of-text marker in the document

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

herer

-0.66

emale

-0.60

 beforehand

-0.59

ornings

-0.58

/"

-0.56

*.

-0.54

ugen

-0.54

 accordingly

-0.54

 afterwards

-0.53

EF

-0.53

POSITIVE LOGITS

 same

0.83

 foregoing

0.80

 hottest

0.77

 entirety

0.75

 remainder

0.75

 simplest

0.74

 largest

0.73

resa

0.72

 smallest

0.71

 following

0.71

Activations Density 0.759%

phrases related to specific actions or descriptions of events occurring

the presence of an end-of-text marker in the document

No Comments

No Known Activations