INDEX

Explanations

specific formatting and layout elements in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anskje

-0.65

 esternos

-0.63

 مشين

-0.61

 ſche

-0.61

 Anſ

-0.61

genodigd

-0.61

 pleaſure

-0.60

InputTagHelper

-0.60

WireFormatLite

-0.59

 ddelwed

-0.59

POSITIVE LOGITS

0.35

und

0.35

0.34

 endblock

0.33

po

0.32

 неба

0.31

0.30

ill

0.30

du

0.30

Activations Density 0.002%

specific formatting and layout elements in code

No Comments

No Known Activations