INDEX

Explanations

proper nouns related to individuals or companies named "Chip"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly those named Chip and Gene

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ahime

-0.70

paces

-0.67

hips

-0.67

adesh

-0.64

 basket

-0.64

ashtra

-0.64

hip

-0.63

 compound

-0.63

ascular

-0.63

egal

-0.63

POSITIVE LOGITS

 Kelly

1.08

 Whitman

0.98

 Snyder

0.97

otle

0.97

 Simmons

0.96

 Warner

0.93

 McKenzie

0.91

 DeVos

0.90

 Smith

0.89

Dug

0.88

Activations Density 0.039%

proper nouns related to individuals or companies named "Chip"

mentions of specific individuals, particularly those named Chip and Gene

No Comments

No Known Activations