INDEX

Explanations

words related to names, likely focusing on last names

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals or authors with the initial "F."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Gemini

-0.72

 enough

-0.61

 ashes

-0.59

 maternity

-0.58

 giants

-0.58

 gent

-0.57

 machine

-0.57

Pv

-0.57

 Passenger

-0.56

 mentally

-0.55

POSITIVE LOGITS

ilipp

1.43

uchs

1.39

ortun

1.38

ergus

1.36

ruits

1.18

ierce

1.15

ultz

1.15

ogle

1.14

ortunate

1.14

irth

1.13

Activations Density 0.029%

words related to names, likely focusing on last names

mentions of specific individuals or authors with the initial "F."

No Comments

No Known Activations