INDEX

Explanations

mentions of the word "Us" with different levels of activation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Us."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

served

-0.88

hetti

-0.72

 chaired

-0.67

 Clarkson

-0.67

*/(

-0.66

 confinement

-0.62

lined

-0.62

paralle

-0.59

aired

-0.58

 secrecy

-0.58

POSITIVE LOGITS

hers

1.11

agi

1.05

ages

0.90

selves

0.87

ern

0.85

uman

0.85

urers

0.85

chwitz

0.82

ports

0.80

ual

0.80

Activations Density 0.014%

mentions of the word "Us" with different levels of activation

occurrences of the word "Us."

No Comments

No Known Activations