INDEX

Explanations

technical terms or jargon used in scientific or professional contexts

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and their professional or personal attributes

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

escription

-0.82

ngth

-0.78

haar

-0.77

GGGGGGGG

-0.77

auts

-0.75

ritis

-0.74

rehensive

-0.72

htaking

-0.70

DragonMagazine

-0.70

lves

-0.69

POSITIVE LOGITS

 occasions

1.20

 behalf

1.16

 basis

1.16

 occasion

0.93

 grounds

0.92

 bandwagon

0.87

 fronts

0.87

 pretext

0.84

ilts

0.83

eve

0.83

Activations Density 0.748%

technical terms or jargon used in scientific or professional contexts

references to specific individuals and their professional or personal attributes

No Comments

No Known Activations