INDEX

Explanations

mentions of confusion and clarity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

secta

-0.47

valuer

-0.46

nasel

-0.46

specialchars

-0.46

@}

-0.44

devamını

-0.44

ofire

-0.44

atchewan

-0.44

Socio

-0.42

WritableDatabase

-0.42

POSITIVE LOGITS

 confusion

2.51

 Confusion

2.14

 confuse

2.12

 confused

2.11

Confusion

2.09

 confusing

2.08

confusion

2.00

 confuses

1.96

 confusión

1.85

 Confused

1.83

Activations Density 0.453%

mentions of confusion and clarity in various contexts

No Comments

No Known Activations