INDEX

Explanations

expressions of agreement

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 guidance

-0.49

 Guidance

-0.47

 Guidelines

-0.46

 GenerationType

-0.44

 guidelines

-0.43

 Statement

-0.42

ori

-0.41

rinfo

-0.40

 Detection

-0.40

Guidance

-0.40

POSITIVE LOGITS

 agree

0.89

 Agree

0.74

 setuju

0.74

agree

0.73

Agree

0.72

 agrees

0.69

 innovación

0.66

 disagree

0.65

 agreeing

0.64

 agreed

0.60

Activations Density 0.121%

expressions of agreement

No Comments

No Known Activations