INDEX

Explanations

complex ideas related to language and meaning-making processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enge

-0.15

 Habit

-0.14

ÙĤØ±Ø§Ø±

-0.14

hab

-0.14

á»īnh

-0.14

imit

-0.13

ç

-0.13

sprintf

-0.13

indow

-0.13

ë²Į

-0.13

POSITIVE LOGITS

 meaning

0.76

 meanings

0.65

meaning

0.60

 Meaning

0.60

-mean

0.43

 signific

0.42

 significance

0.41

Mean

0.40

æĦıä¹ī

0.37

 interpretation

0.36

Activations Density 0.438%

complex ideas related to language and meaning-making processes

No Comments

No Known Activations