INDEX

Explanations

translations or references to translated materials

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "translated" and variations of translation-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

achine

-0.79

Peb

-0.76

poral

-0.73

dinand

-0.70

etheus

-0.69

emetery

-0.69

ucket

-0.69

 Gamble

-0.69

--+

-0.68

POSITIVE LOGITS

 translations

0.96

 translation

0.96

 translator

0.92

translation

0.87

 into

0.86

 transl

0.85

 subtitles

0.84

 translated

0.83

into

0.82

lator

0.81

Activations Density 0.043%

translations or references to translated materials

occurrences of the word "translated" and variations of translation-related terminology

No Comments

No Known Activations