INDEX

Explanations

mentions of controversial or disputed assertions or allegations

oai_token-act-pair · gpt-3.5-turbo

references to claims or assertions made in the text

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

 newcom

-0.76

 arrang

-0.74

 Watching

-0.72

srf

-0.70

ascript

-0.68

 guiActiveUnfocused

-0.67

etheless

-0.66

otos

-0.65

 unpre

-0.64

matically

-0.63

POSITIVE LOGITS

 claims

1.01

ylum

0.88

Claim

0.85

ifications

0.79

 Claims

0.78

urances

0.77

icist

0.77

 claim

0.76

pring

0.76

 assertions

0.75

Activations Density 0.021%

mentions of controversial or disputed assertions or allegations

references to claims or assertions made in the text

No Comments

No Known Activations