INDEX

Explanations

phrases related to actions or commands

oai_token-act-pair · gpt-3.5-turbo

action words that indicate gameplay and interaction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cius

-0.71

thodox

-0.67

 unlaw

-0.67

aith

-0.64

icum

-0.64

rake

-0.63

Applic

-0.63

minecraft

-0.61

LGBT

-0.60

iosyn

-0.60

POSITIVE LOGITS

 yourself

1.04

 yourselves

0.97

 your

0.83

Tube

0.65

animate

0.61

 butterflies

0.61

 YOUR

0.60

lda

0.60

pez

0.58

 realise

0.58

Activations Density 0.190%

phrases related to actions or commands

action words that indicate gameplay and interaction

No Comments

No Known Activations