INDEX

Explanations

words related to protests and protection

oai_token-act-pair · gpt-3.5-turbo

references to various types of protections or protocols

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

manship

-0.85

INESS

-0.82

uliffe

-0.79

íķ

-0.79

=-=-=-=-

-0.73

è£ħ

-0.73

aneers

-0.72

 mund

-0.72

LINE

-0.71

hips

-0.71

POSITIVE LOGITS

otyp

1.17

agonist

1.17

ected

1.15

otypes

1.14

ests

1.08

ocol

0.97

 Prot

0.96

agonists

0.93

otype

0.87

racted

0.82

Activations Density 0.013%

words related to protests and protection

references to various types of protections or protocols

No Comments

No Known Activations