INDEX

Explanations

terms related to geographical or conceptual zones

oai_token-act-pair · gpt-3.5-turbo

references to distinct geographical or operational areas designated as "zones."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

anooga

-0.83

bats

-0.81

issance

-0.80

utive

-0.79

thora

-0.77

ndra

-0.76

lycer

-0.72

omore

-0.72

ever

-0.68

ÄŁ

-0.68

POSITIVE LOGITS

 boundary

0.97

 zones

0.97

 zone

0.95

 boundaries

0.95

 deline

0.95

 dwell

0.85

 area

0.82

 Areas

0.82

inh

0.81

 separating

0.81

Activations Density 0.054%

terms related to geographical or conceptual zones

references to distinct geographical or operational areas designated as "zones."

No Comments

No Known Activations