INDEX

Explanations

phrases indicating something is under consideration or influence

oai_token-act-pair · gpt-3.5-turbo

instances of the word "subject" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

yip

-0.67

ERROR

-0.64

cia

-0.61

owler

-0.59

ces

-0.58

ttp

-0.57

hetto

-0.56

SER

-0.55

 cane

-0.55

gian

-0.54

POSITIVE LOGITS

ively

1.38

ivity

1.33

ivist

1.23

ivities

1.15

matter

1.09

ivism

1.09

ive

1.08

ion

1.05

ivation

1.02

ivating

0.99

Activations Density 0.032%

phrases indicating something is under consideration or influence

instances of the word "subject" in various contexts

No Comments

No Known Activations