INDEX

Explanations

repetitions of the word "EqualsAnd" and possibly related to the structure of statistical summaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

thâu

-0.62

 purpoſe

-0.61

onnay

-0.60

 himſelf

-0.60

 ſec

-0.58

 themſelves

-0.57

guien

-0.56

 uſed

-0.56

الإنجليزية

-0.55

 firſt

-0.55

POSITIVE LOGITS

And

1.11

và

1.07

Và

1.02

 και

1.02

그리고

1.02

\&

1.01

AND

1.01

и

0.97

和

0.95

AndGet

0.95

Activations Density 0.146%

repetitions of the word "EqualsAnd" and possibly related to the structure of statistical summaries

No Comments

No Known Activations