INDEX

Explanations

fun activities or experiences

oai_token-act-pair · gpt-3.5-turbo

expressions that convey enjoyment or fun activities

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

ledged

-0.81

avering

-0.73

 authorised

-0.70

vation

-0.69

pex

-0.69

 cleared

-0.69

 mandated

-0.68

insured

-0.68

 liability

-0.67

igham

-0.67

POSITIVE LOGITS

GIF

0.92

 trivia

0.88

gif

0.83

 Pixar

0.77

 Animation

0.74

Reviewer

0.74

yrics

0.74

ãĤ¼

0.74

 Mahjong

0.70

nesday

0.70

Activations Density 0.363%

fun activities or experiences

expressions that convey enjoyment or fun activities

No Comments

No Known Activations