INDEX

Explanations

descriptions of vulnerability and emotional states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Monfieur

-0.93

 himſelf

-0.92

 myſelf

-0.91

 chofe

-0.89

pexpr

-0.87

 Houſe

-0.86

Eſ

-0.84

 ſtand

-0.84

 ſever

-0.84

Efq

-0.83

POSITIVE LOGITS

رشف

0.46

ra

0.43

 summers

0.40

im

0.40

rad

0.40

tenberg

0.40

 بگو

0.40

Im

0.39

 Perhaps

0.38

0.37

Activations Density 0.313%

descriptions of vulnerability and emotional states

No Comments

No Known Activations