INDEX

Explanations

references to the word "Har" with varying levels of intensity (e.g., Har, Har 10, Har 9)

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or name, particularly "Har"

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

éĹĺ

-0.87

 Tone

-0.83

xual

-0.66

REDACTED

-0.63

UCT

-0.63

 semantics

-0.63

anwhile

-0.62

ĸļ

-0.61

eers

-0.61

 sheet

-0.61

POSITIVE LOGITS

assment

1.27

vard

1.08

poons

1.06

rier

1.05

riers

1.04

poon

1.04

rod

1.02

bors

1.00

rah

1.00

assed

0.99

Activations Density 0.032%

references to the word "Har" with varying levels of intensity (e.g., Har, Har 10, Har 9)

references to a specific entity or name, particularly "Har"

No Comments

No Known Activations