INDEX

Explanations

terms related to candidates and elections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/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.17.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

 jadx

-0.60

 départements

-0.59

Skocz

-0.59

hoe

-0.57

roads

-0.55

Hrsg

-0.54

worths

-0.54

 Lumpur

-0.53

roach

-0.53

writerow

-0.52

POSITIVE LOGITS

CAND

1.01

 CAND

0.95

 ModelExpression

0.95

RenderAtEndOf

0.94

 Candidates

0.94

 Cand

0.91

 Monfieur

0.91

 Majefty

0.90

 candidates

0.90

 candidat

0.86

Activations Density 0.262%

terms related to candidates and elections

No Comments

No Known Activations