INDEX

Explanations

mentions of actions or events that could potentially happen

oai_token-act-pair · gpt-3.5-turbo

expressions of uncertainty or possibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Rohingya

-0.63

 Syrian

-0.60

 anthem

-0.57

tained

-0.56

ussion

-0.56

 indigenous

-0.54

 Sudan

-0.54

BuyableInstoreAndOnline

-0.53

 Headquarters

-0.53

 Mali

-0.53

POSITIVE LOGITS

 certainly

0.92

unless

0.85

 wouldn

0.84

 preferably

0.80

ichever

0.76

fortunately

0.76

etheless

0.74

 fortunately

0.74

assuming

0.74

depending

0.73

Activations Density 1.034%

mentions of actions or events that could potentially happen

expressions of uncertainty or possibility

No Comments

No Known Activations