INDEX

Explanations

phrases indicating a change in position or stance in a conversation

oai_token-act-pair · gemini-2.0-flash

Evasion

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RegressionTest

-0.88

PerformLayout

-0.85

 متعلقه

-0.82

 BoxFit

-0.68

rizona

-0.66

NewLabel

-0.65

 resourceCulture

-0.63

참고

-0.62

UrlResolution

-0.62

AnimationsModule

-0.62

POSITIVE LOGITS

ev

0.66

 dodge

0.61

mum

0.58

 diplom

0.58

 evasion

0.57

 dodged

0.57

 dodging

0.57

 diplomatic

0.56

 Tuch

0.55

 skirting

0.53

Activations Density 3.588%

phrases indicating a change in position or stance in a conversation

Evasion

No Comments

No Known Activations