INDEX

Explanations

discussions of bias and opinion in political contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 здра

-0.40

슷

-0.38

InitStruct

-0.38

stanze

-0.38

 compét

-0.37

engertian

-0.36

PullParser

-0.36

Rozm

-0.36

handleDelete

-0.35

tá

-0.35

POSITIVE LOGITS

 biased

0.60

 bias

0.56

 prejudiced

0.55

bias

0.54

biased

0.54

новништво

0.52

Bias

0.52

 neutrality

0.50

 biases

0.50

biases

0.50

Activations Density 1.111%

discussions of bias and opinion in political contexts

No Comments

No Known Activations