INDEX

Explanations

translations or mentions of translation activities

oai_token-act-pair · gpt-3.5-turbo

references to translation activities or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ndra

-0.89

hattan

-0.76

pton

-0.76

achine

-0.76

--+

-0.75

ategory

-0.75

ailability

-0.74

negie

-0.72

etheus

-0.72

levard

-0.71

POSITIVE LOGITS

 into

1.11

into

1.06

 translations

1.04

 subtitles

1.00

 translator

0.95

 translation

0.94

 Into

0.92

 INTO

0.86

 translated

0.85

lations

0.84

Activations Density 0.062%

translations or mentions of translation activities

references to translation activities or processes

No Comments

No Known Activations