INDEX

Explanations

references to lost wages and damages in legal 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

硃

-0.41

 Sepp

-0.39

TabStop

-0.38

WithIOException

-0.37

 McGee

-0.37

 olsa

-0.36

chid

-0.36

tick

-0.36

 اسک

-0.36

 flat

-0.35

POSITIVE LOGITS

 purpoſe

0.63

 auroit

0.62

 feroit

0.62

 myſelf

0.59

 ſever

0.58

 CURIAM

0.57

 themſelves

0.57

CDCl

0.57

 perſon

0.57

 pleaſure

0.57

Activations Density 0.046%

references to lost wages and damages in legal contexts

No Comments

No Known Activations