INDEX

Explanations

locations

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to places and establishments

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

ictions

-0.87

ought

-0.86

amy

-0.85

iction

-0.81

kowski

-0.77

ledged

-0.77

lasses

-0.76

stories

-0.76

employed

-0.76

essage

-0.75

POSITIVE LOGITS

de

1.19

 Grande

1.19

 Royale

1.16

del

1.15

 Palest

1.14

du

1.12

 Parad

1.12

Vec

1.10

 Prin

1.08

 Libre

1.06

Activations Density 0.195%

locations

proper nouns related to places and establishments

No Comments

No Known Activations