INDEX

Explanations

words related to controversial claims or actions

oai_token-act-pair · gpt-3.5-turbo

negatively charged or dismissive language related to claims and actions

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

 Pirates

-0.70

 Ninja

-0.65

Topics

-0.64

 Bronze

-0.63

 Commando

-0.62

 Reds

-0.62

 Engineers

-0.61

HC

-0.60

 Predators

-0.60

BuyableInstoreAndOnline

-0.60

POSITIVE LOGITS

 dearly

0.90

 aloud

0.84

 anyways

0.82

 beforehand

0.79

.</

0.79

 firsthand

0.77

 anyway

0.77

.<

0.74

 enthusiastically

0.74

 proudly

0.72

Activations Density 0.232%

words related to controversial claims or actions

negatively charged or dismissive language related to claims and actions

No Comments

No Known Activations