INDEX

Explanations

terms related to computer technology, specifically focusing on security sandbox methods

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to methods or techniques for self-improvement or education

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

 Mits

-0.80

Av

-0.80

Isa

-0.78

Isa

-0.76

Lu

-0.73

eteenth

-0.71

flu

-0.69

Lu

-0.68

ighth

-0.67

ó

-0.66

POSITIVE LOGITS

Sand

2.39

 Sand

2.33

 Rock

1.94

 sand

1.88

sand

1.86

Rock

1.84

 Rocks

1.59

rock

1.58

 ROCK

1.57

 rock

1.52

Activations Density 0.360%

terms related to computer technology, specifically focusing on security sandbox methods

references to methods or techniques for self-improvement or education

No Comments

No Known Activations