INDEX

Explanations

quantitative measurements such as percentages and numbers

oai_token-act-pair · gpt-3.5-turbo

specific numerical values and physical performance metrics

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Berks

-0.78

john

-0.76

ãĤ»

-0.75

©¶æ

-0.75

 Jaguar

-0.74

ĳå£«

-0.72

Comb

-0.70

razil

-0.69

compan

-0.68

otos

-0.66

POSITIVE LOGITS

 velocity

1.13

 elevation

1.10

 voltage

1.10

 length

1.10

 size

1.07

 potency

1.06

 rate

1.05

 value

1.05

 speed

1.02

 magnitude

1.02

Activations Density 0.365%

quantitative measurements such as percentages and numbers

specific numerical values and physical performance metrics

No Comments

No Known Activations