INDEX

Explanations

occurrences of specific proper nouns or formal titles

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

 محفوظة

-0.57

 régler

-0.45

外部リンク

-0.45

 kepentingan

-0.45

 kepem

-0.44

 المعيارى

-0.43

 الوطنيه

-0.42

 üzere

-0.42

WriteTagHelper

-0.42

 čin

-0.42

POSITIVE LOGITS

 said

1.03

 explained

0.93

 says

0.90

ThroughAttribute

0.85

 dijo

0.84

 told

0.83

 commented

0.82

 recalled

0.82

said

0.81

 spokeswoman

0.80

Activations Density 0.123%

occurrences of specific proper nouns or formal titles

No Comments

No Known Activations