INDEX

Explanations

words related to negative public reaction or controversy

oai_token-act-pair · gpt-3.5-turbo

references to controversies and public backlash

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

tein

-0.99

istries

-0.69

nesota

-0.69

 subsistence

-0.68

 planner

-0.64

auga

-0.62

sonian

-0.61

ete

-0.61

elsen

-0.61

 parole

-0.59

POSITIVE LOGITS

 surrounding

1.10

 stemming

0.90

 regarding

0.89

 ensued

0.86

rovers

0.86

 arising

0.85

 pertaining

0.82

 caused

0.82

 generated

0.79

 uproar

0.76

Activations Density 0.130%

words related to negative public reaction or controversy

references to controversies and public backlash

No Comments

No Known Activations