INDEX

Explanations

terms related to academic departments or units

oai_token-act-pair · gpt-3.5-turbo

mentions of various departments or governmental organizations

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

raise

-0.63

shot

-0.61

milo

-0.60

 largeDownload

-0.60

 proto

-0.60

 âľĵ

-0.60

ĻĤ

-0.59

hammad

-0.58

ael

-0.58

casts

-0.57

POSITIVE LOGITS

 department

3.87

 departments

2.88

 Department

2.25

Department

2.20

 Dept

2.04

 bureau

1.62

 ministry

1.42

 agency

1.38

artment

1.23

artments

1.22

Activations Density 0.009%

terms related to academic departments or units

mentions of various departments or governmental organizations

No Comments

No Known Activations