INDEX

Explanations

text that mention translations or translated content

oai_token-act-pair · gpt-3.5-turbo

references to translations of texts or media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ndra

-1.01

achine

-0.76

Peb

-0.75

pload

-0.73

 Gamble

-0.72

reditary

-0.72

resy

-0.70

dinand

-0.70

\/\/

-0.70

pton

-0.70

POSITIVE LOGITS

 translation

1.03

 translations

1.02

 translator

0.94

 transl

0.92

 translated

0.91

translation

0.89

 into

0.86

 subtitles

0.85

 translates

0.83

language

0.81

Activations Density 0.039%

text that mention translations or translated content

references to translations of texts or media

No Comments

No Known Activations