INDEX

Explanations

text related to logos

oai_token-act-pair · gpt-3.5-turbo

references to logos and branding visuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Lank

-0.72

rough

-0.68

astern

-0.67

theless

-0.66

Ago

-0.63

artz

-0.63

tein

-0.63

uth

-0.62

hurst

-0.61

 Lanka

-0.61

POSITIVE LOGITS

 logo

1.07

 logos

0.98

 Logo

0.96

 Doodle

0.93

 Seym

0.81

 slogan

0.81

matic

0.79

matically

0.76

 emblem

0.76

 flare

0.75

Activations Density 0.017%

text related to logos

references to logos and branding visuals

No Comments

No Known Activations