INDEX

Explanations

sequences related to transcription processes

oai_token-act-pair · gpt-3.5-turbo

terms related to transcription and publication processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

erva

-0.80

EStream

-0.74

HMS

-0.72

 Lumpur

-0.72

emort

-0.69

¬¼

-0.67

agle

-0.66

 ItemLevel

-0.66

enger

-0.64

ghazi

-0.64

POSITIVE LOGITS

 transcription

1.05

kefeller

0.91

 photoc

0.81

riter

0.77

ingen

0.77

sis

0.76

 plur

0.75

 corrid

0.74

hiba

0.74

chrome

0.73

Activations Density 0.027%

sequences related to transcription processes

terms related to transcription and publication processes

No Comments

No Known Activations