INDEX

Explanations

locations or landmarks

oai_token-act-pair · gpt-3.5-turbo

place names and cities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ICES

-0.80

inventory

-0.79

ledged

-0.79

acting

-0.77

slot

-0.77

responsible

-0.74

peria

-0.72

reviewed

-0.72

vol

-0.72

hement

-0.71

POSITIVE LOGITS

 skyline

1.20

 Nights

1.08

 natives

1.04

 Skies

1.00

 winters

0.90

 resident

0.88

 Massacre

0.88

 Lights

0.87

 Heist

0.86

 Scene

0.85

Activations Density 0.405%

locations or landmarks

place names and cities

No Comments

No Known Activations