INDEX

Explanations

terms related to the occurrence and frequency of events or conditions

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

statistical phrases involving the word "occurrence".

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 Anſ

-0.68

 Мексичка

-0.64

tshell

-0.60

GHIJKLM

-0.59

wiſe

-0.57

saveButton

-0.57

Thin

-0.55

SAVE

-0.55

Tapi

-0.54

 laſt

-0.54

POSITIVE LOGITS

 occurrence

1.17

 occur

1.02

 occurrences

1.00

 occurring

0.98

 Occur

0.97

 occurs

0.93

occur

0.90

 occurred

0.90

 Occurrence

0.89

occurrence

0.83

Activations Density 0.159%

terms related to the occurrence and frequency of events or conditions

statistical phrases involving the word "occurrence".

No Comments

No Known Activations