INDEX

Explanations

solutions related to measurements and quantities

oai_token-act-pair · gpt-3.5-turbo

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

 tender

-0.81

 poaching

-0.77

 bonuses

-0.74

 nods

-0.73

 unmist

-0.73

 mimic

-0.72

 amounts

-0.71

 interchange

-0.70

 disappearing

-0.69

 notices

-0.69

POSITIVE LOGITS

each

1.21

dL

1.19

1.16

hr

1.16

hour

1.14

sq

1.13

month

1.11

etc

1.09

lbs

1.09

rev

1.06

Activations Density 0.032%

solutions related to measurements and quantities

No Comments

No Known Activations