INDEX

Explanations

terms related to high-profile individuals or entities

oai_token-act-pair · gpt-3.5-turbo

references to high-profile individuals or cases

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

shall

-0.76

Oo

-0.74

otent

-0.73

neg

-0.70

aves

-0.69

uddled

-0.68

agan

-0.66

agree

-0.64

ften

-0.64

ogie

-0.62

POSITIVE LOGITS

profile

1.02

 profiles

0.99

 profile

0.88

Profile

0.86

allery

0.80

 è£ıè

0.78

 engagements

0.75

 campaigns

0.73

 celebrities

0.73

 headlines

0.72

Activations Density 0.010%

terms related to high-profile individuals or entities

references to high-profile individuals or cases

No Comments

No Known Activations