INDEX

Explanations

phrases related to specific place names, possibly indicating geographical locations or organizations

oai_token-act-pair · gpt-3.5-turbo

specific names and places related to governmental or community-focused initiatives

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

afort

-0.82

sounding

-0.66

THING

-0.66

epad

-0.65

ãģĨ

-0.64

oshop

-0.63

zsche

-0.62

forcer

-0.61

 Bane

-0.61

auntlet

-0.60

POSITIVE LOGITS

 users

1.50

 residents

1.49

 learners

1.44

 students

1.43

 travelers

1.41

 lovers

1.40

 citizens

1.37

 consumers

1.36

 subscribers

1.36

 retirees

1.36

Activations Density 0.702%

phrases related to specific place names, possibly indicating geographical locations or organizations

specific names and places related to governmental or community-focused initiatives

No Comments

No Known Activations