INDEX

Explanations

numbers and related mathematical operations

oai_token-act-pair · gpt-3.5-turbo

numerical sequences or values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

termin

-0.70

killed

-0.70

orc

-0.69

lethal

-0.68

hell

-0.67

phe

-0.67

Cro

-0.65

Buzz

-0.64

ones

-0.63

Sov

-0.63

POSITIVE LOGITS

 <!--

0.92

 Â¯

0.87

 =================================================================

0.84

 Flavoring

0.79

 Submit

0.77

 NOTICE

0.76

 Next

0.76

:=

0.76

 âĸ

0.73

 TEAM

0.73

Activations Density 0.016%

numbers and related mathematical operations

numerical sequences or values

No Comments

No Known Activations