INDEX

Explanations

technology-related terms, specifically related to computers, programming, and software development

oai_token-act-pair · gpt-3.5-turbo

symbols and punctuation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

icum

-0.83

ord

-0.81

oresc

-0.78

ixel

-0.77

asma

-0.77

ibility

-0.76

oit

-0.74

flies

-0.72

odi

-0.71

kinson

-0.70

POSITIVE LOGITS

âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦

1.08

âĢ¦âĢ¦âĢ¦âĢ¦

1.04

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

0.88

////////////////////////////////

0.83

.....

0.71

 TRANS

0.70

————————

0.68

 Continued

0.66

BUT

0.65

CLASSIFIED

0.65

Activations Density 0.030%

technology-related terms, specifically related to computers, programming, and software development

symbols and punctuation

No Comments

No Known Activations