INDEX

Explanations

names of individuals being selected or hired for specific roles or positions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "to."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 pauses

-0.78

grain

-0.77

 glances

-0.75

artifacts

-0.69

gran

-0.68

tions

-0.66

browser

-0.66

pointers

-0.64

egg

-0.62

noticed

-0.62

POSITIVE LOGITS

 participate

1.14

 join

1.01

 defend

1.00

 oversee

1.00

 compensate

0.99

 fulfill

0.99

 provide

0.98

 give

0.97

 meet

0.97

 negotiate

0.97

Activations Density 0.216%

names of individuals being selected or hired for specific roles or positions

instances of the word "to."

No Comments

No Known Activations