INDEX

Explanations

references to awards or accolades received by individuals or works

oai_token-act-pair · gpt-3.5-turbo

mentions of various awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Geo

-0.69

ket

-0.69

ickr

-0.65

alter

-0.64

loo

-0.64

gun

-0.63

abol

-0.62

 alter

-0.61

okin

-0.61

sen

-0.60

POSITIVE LOGITS

 Award

1.45

 Awards

1.33

 award

1.08

 awards

1.07

 Winners

1.03

Winner

1.02

 Prize

1.02

 awarded

0.99

 Trophy

0.95

 laure

0.90

Activations Density 0.010%

references to awards or accolades received by individuals or works

mentions of various awards

No Comments

No Known Activations