INDEX

Explanations

mentions of the word "ott" at varying activations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Ott."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

senal

-0.72

 calculus

-0.65

 reserve

-0.64

 Gemini

-0.63

 CONTROL

-0.61

 prosecution

-0.60

eer

-0.60

ggle

-0.60

 spreadsheet

-0.60

 coordinate

-0.60

POSITIVE LOGITS

ott

1.33

sburgh

1.03

awa

0.92

stown

0.90

terness

0.89

ito

0.89

orn

0.89

ido

0.88

itt

0.87

eca

0.87

Activations Density 0.004%

mentions of the word "ott" at varying activations

occurrences of the name "Ott."

No Comments

No Known Activations