INDEX

Explanations

concepts related to positivity and beneficial outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

breviations

-0.47

áctenos

-0.47

ophobia

-0.45

telles

-0.45

scrollTo

-0.44

 őket

-0.44

-0.43

mjs

-0.42

Itinerary

-0.42

chond

-0.42

POSITIVE LOGITS

 Positive

0.98

Positive

0.97

 positive

0.97

Negative

0.93

 Negative

0.90

positive

0.90

 POSITIVE

0.89

POSITIVE

0.88

 negative

0.85

negative

0.81

Activations Density 0.229%

concepts related to positivity and beneficial outcomes

No Comments

No Known Activations