INDEX

Explanations

references to lips or lip-related actions

oai_token-act-pair · gpt-3.5-turbo

references to lips and lip-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Leap

-0.79

 Newsp

-0.72

ENCY

-0.72

 Schr

-0.68

Ùİ

-0.67

 Reloaded

-0.67

ADRA

-0.66

 Telecommunications

-0.65

 Danger

-0.65

Øª

-0.64

POSITIVE LOGITS

lip

1.06

atures

1.01

ids

0.99

artisan

0.94

oca

0.92

lip

0.92

etsk

0.91

pha

0.86

sticks

0.86

seys

0.85

Activations Density 0.008%

references to lips or lip-related actions

references to lips and lip-related terminology

No Comments

No Known Activations