INDEX

Explanations

references to specific places or names, particularly focusing on the name "Monte"

oai_token-act-pair · gpt-3.5-turbo

references to the figure "Monte" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

inventoryQuantity

-0.81

irlf

-0.81

hips

-0.74

inyl

-0.73

inning

-0.71

ivities

-0.70

utenberg

-0.70

¿

-0.69

hip

-0.67

Marginal

-0.66

POSITIVE LOGITS

 Monte

1.05

 Carlo

1.04

fort

0.87

ucci

0.85

lette

0.81

erey

0.78

ecast

0.76

leys

0.75

 Blanc

0.75

 Crist

0.74

Activations Density 0.007%

references to specific places or names, particularly focusing on the name "Monte"

references to the figure "Monte" in various contexts

No Comments

No Known Activations