INDEX

Explanations

words preceding a hyphen that indicate a previous status or occupation

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "ex-" indicating something that was previously in a role or relationship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

kson

-0.91

atche

-0.77

zsche

-0.76

 Twain

-0.72

oglu

-0.66

 Meadows

-0.66

DeV

-0.66

 Pryor

-0.65

izont

-0.65

Tac

-0.65

POSITIVE LOGITS

girlfriend

1.00

sent

0.99

favorite

0.96

husband

0.91

chance

0.90

turned

0.90

vict

0.89

training

0.89

national

0.88

decl

0.87

Activations Density 0.108%

words preceding a hyphen that indicate a previous status or occupation

instances of the prefix "ex-" indicating something that was previously in a role or relationship

No Comments

No Known Activations