INDEX

Explanations

words related to discussion or analysis about a particular topic or piece of media

oai_token-act-pair · gpt-3.5-turbo

phrases indicating critical discussion or analysis of media content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

jri

-0.82

elong

-0.70

urdue

-0.69

cius

-0.68

ollah

-0.66

reprene

-0.63

urers

-0.62

yip

-0.62

etooth

-0.61

LinkedIn

-0.61

POSITIVE LOGITS

 Episode

1.06

 Tales

1.04

 Shattered

1.02

 Into

1.02

 Scream

1.02

 Revenge

1.01

 Unch

1.00

 Season

0.99

 Labyrinth

0.95

 Interstellar

0.94

Activations Density 0.366%

words related to discussion or analysis about a particular topic or piece of media

phrases indicating critical discussion or analysis of media content

No Comments

No Known Activations