INDEX

Explanations

the word "librarian."

oai_token-act-pair · gpt-3.5-turbo

mentions of libraries and related terminology

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

 sake

-0.70

 potato

-0.70

 fallacy

-0.69

 avoidance

-0.68

 cone

-0.67

 passes

-0.67

 pony

-0.65

 rake

-0.65

 wrap

-0.64

 Cowboys

-0.63

POSITIVE LOGITS

ibr

4.65

ibrarian

1.55

ib

1.45

iber

1.25

ocal

1.16

ibe

1.16

ibl

1.14

igor

1.12

usc

1.11

ilit

1.08

Activations Density 0.017%

the word "librarian."

mentions of libraries and related terminology

No Comments

No Known Activations