INDEX

Explanations

mentions of specific TV networks or channels

oai_token-act-pair · gpt-3.5-turbo

mentions of specific television networks and military organizations

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

venge

-0.77

eve

-0.77

pired

-0.73

emy

-0.71

¥µ

-0.70

leans

-0.67

Shot

-0.67

 Schultz

-0.65

essor

-0.64

rollment

-0.64

POSITIVE LOGITS

Â£

1.00

ITV

1.00

RAF

0.93

 Buckingham

0.93

 Surrey

0.91

BST

0.89

Britain

0.85

 Britain

0.84

NHS

0.81

 Â£

0.80

Activations Density 0.036%

mentions of specific TV networks or channels

mentions of specific television networks and military organizations

No Comments

No Known Activations