INDEX

Explanations

places or actions associated with various activities

oai_token-act-pair · gpt-3.5-turbo

places associated with specific actions or functions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

requires

-0.74

UTH

-0.69

 warrants

-0.66

 RELE

-0.66

 VERS

-0.62

 batches

-0.61

Ole

-0.60

CSV

-0.58

 Parallel

-0.58

 cooker

-0.57

POSITIVE LOGITS

 dwell

0.83

veyard

0.77

ocate

0.77

ezvous

0.75

rongh

0.74

isphere

0.73

aign

0.73

anwhile

0.73

live

0.72

iage

0.72

Activations Density 0.088%

places or actions associated with various activities

places associated with specific actions or functions

No Comments

No Known Activations