INDEX

Explanations

key phrases indicating comparisons or contrasts, especially relating to statistics or social groups

oai_token-act-pair · gpt-3.5-turbo

phrases related to comparative group dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

interrupted

-0.71

ens

-0.67

 nailed

-0.63

ymes

-0.60

 sacked

-0.60

 disqualified

-0.57

 extraord

-0.57

 lodged

-0.57

 encouraged

-0.57

tab

-0.57

POSITIVE LOGITS

 regards

0.83

 regard

0.81

roads

0.80

versions

0.80

Demand

0.71

 Nanto

0.70

ãĥ¼ãĥĨ

0.68

clusively

0.67

eff

0.67

 spite

0.66

Activations Density 0.662%

key phrases indicating comparisons or contrasts, especially relating to statistics or social groups

phrases related to comparative group dynamics

No Comments

No Known Activations