INDEX

Explanations

references to servicemen or related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to service members or veterans

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

cob

-0.68

br

-0.66

 corporate

-0.58

HIP

-0.58

 Cance

-0.58

Big

-0.58

sp

-0.58

 spiral

-0.57

 insurance

-0.57

 disappro

-0.57

POSITIVE LOGITS

emen

4.99

eman

2.32

amen

1.44

emin

1.22

emon

1.14

eland

1.12

emark

1.12

emet

1.11

MEN

1.10

men

1.10

Activations Density 0.012%

references to servicemen or related terms

terms related to service members or veterans

No Comments

No Known Activations