INDEX

Explanations

expressions of appreciation and positivity

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

arpa

-0.46

 καν

-0.36

CppMethod

-0.36

Jîn

-0.35

 Alphabet

-0.34

 context

-0.34

akhstan

-0.33

 Targ

-0.33

╣

-0.33

nament

-0.33

POSITIVE LOGITS

 OMITBAD

0.53

 виправивши

0.52

localctx

0.52

 Савезне

0.51

 snippetHide

0.51

writeFieldEnd

0.47

OGND

0.45

 saites

0.45

'{@

0.45

IUrlHelper

0.44

Activations Density 0.004%

expressions of appreciation and positivity

No Comments

No Known Activations