INDEX

Explanations

terms related to governmental departments or agencies

oai_token-act-pair · gpt-3.5-turbo

references to U.S. government departments and agencies

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

ilant

-0.71

jad

-0.71

leys

-0.67

plets

-0.65

 emph

-0.65

GW

-0.62

gest

-0.61

innacle

-0.60

ãĥŃ

-0.60

gling

-0.59

POSITIVE LOGITS

 Commissioners

0.76

 Instruction

0.74

 Merrill

0.73

ĳå£«

0.71

 Examination

0.69

 Appropriations

0.69

 Education

0.64

 Administration

0.64

 Reconstruction

0.63

 Appeals

0.63

Activations Density 0.086%

terms related to governmental departments or agencies

references to U.S. government departments and agencies

No Comments

No Known Activations