INDEX

Explanations

words related to specific names and locations related to educational institutions and individuals

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those associated with the Hayden Planetarium

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

alions

-0.72

ramid

-0.71

icably

-0.70

iment

-0.70

vertis

-0.70

NetMessage

-0.69

berman

-0.68

iments

-0.68

 Antar

-0.67

ammy

-0.67

POSITIVE LOGITS

 Christensen

0.85

 Hayden

0.77

burgh

0.70

 Watts

0.69

renheit

0.68

bryce

0.65

coe

0.65

 Factor

0.64

 Stevens

0.64

Activations Density 0.010%

words related to specific names and locations related to educational institutions and individuals

references to specific individuals, particularly those associated with the Hayden Planetarium

No Comments

No Known Activations