INDEX

Explanations

terms associated with legal actions and wrongful accusations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ConstraintMaker

-0.47

)?;

-0.45

>";

-0.44

 Keuangan

-0.44

"]];

-0.44

 reveló

-0.43

 हूं

-0.41

kor

-0.41

 angiogenesis

-0.40

shows

-0.40

POSITIVE LOGITS

 unjustly

0.69

 needlessly

0.68

 unnecessarily

0.64

 NUKAT

0.60

ThroughAttribute

0.57

fiore

0.57

setVerticalGroup

0.57

 getStart

0.56

NCIA

0.55

 unjust

0.55

Activations Density 0.836%

terms associated with legal actions and wrongful accusations

No Comments

No Known Activations