INDEX

Explanations

text making conclusions based on research or observations

oai_token-act-pair · gemini-2.0-flash

Formula/Findings

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tvguidetime

-1.42

endphp

-1.33

 ddelweddau

-1.21

parsedMessage

-1.13

endpush

-1.13

TemporalType

-1.13

UserScript

-1.11

MLLoader

-1.09

DoubleQuotes

-1.09

migrationBuilder

-1.07

POSITIVE LOGITS

↵↵

1.10

↵

0.96

0.91

<bos>

0.81

holdet

0.78

 overras

0.74

0.71

The

0.71

 näytte

0.69

 dårlig

0.69

Activations Density 2.851%

text making conclusions based on research or observations

Formula/Findings

No Comments

No Known Activations