INDEX

Explanations

the word "aku" with varying intensity levels

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "aku."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Granger

-0.79

 Cognitive

-0.70

 miscarriage

-0.68

 SERV

-0.67

 illusions

-0.66

sson

-0.64

 apprehension

-0.63

Fib

-0.63

alities

-0.63

 Obesity

-0.62

POSITIVE LOGITS

aku

1.18

ichi

1.06

yu

1.01

zeb

0.99

uchi

0.88

pei

0.84

ji

0.84

za

0.83

pport

0.81

ogo

0.81

Activations Density 0.007%

the word "aku" with varying intensity levels

occurrences of the word "aku."

No Comments

No Known Activations