INDEX

Explanations

references to increased levels or high prevalence of certain conditions or factors in various contexts

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

Str

-0.41

WireFormatLite

-0.37

rela

-0.36

Str

-0.36

|}{\

-0.36

via

-0.35

 privately

-0.35

by

-0.35

Val

-0.34

 into

-0.34

POSITIVE LOGITS

Rüyada

0.65

Absence

0.56

 كومونز

0.54

 Üniversitesi

0.54

 विश्वसनीयता

0.54

 ausencia

0.51

ůli

0.51

SequentialGroup

0.51

 absence

0.50

 Absence

0.50

Activations Density 0.079%

references to increased levels or high prevalence of certain conditions or factors in various contexts

No Comments

No Known Activations