INDEX

Explanations

references to filenames or file-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

\[

-0.39

 pétrole

-0.37

‏‏

-0.36

("$.

-0.35

 Audiodateien

-0.35

↩

-0.35

}:${

-0.35

ői

-0.34

 ſur

-0.34

(.*

-0.33

POSITIVE LOGITS

filename

0.85

 filename

0.84

awtextra

0.65

Filename

0.63

InjectAttribute

0.61

von

0.60

 Filename

0.57

RotationOrder

0.56

FieldBuilder

0.55

 '\\;'

0.55

Activations Density 0.071%

references to filenames or file-related terms

No Comments

No Known Activations

references to filenames or file-related terms

No Comments

No Known Activations