INDEX

Explanations

occurrences of the word "all" and references to numbers or groups in multiple contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 overall

-0.17

aga

-0.16

ÑĮÐºÐ¾

-0.15

yna

-0.15

overall

-0.14

 patron

-0.14

anga

-0.14

eka

-0.14

/GPL

-0.13

 rosa

-0.13

POSITIVE LOGITS

ukkit

0.15

0.14

Ung

0.14

åĦĢ

0.14

 corners

0.14

æĸ¹éĿ¢

0.14

axy

0.14

atoi

0.13

Cancelable

0.13

Activations Density 0.087%

occurrences of the word "all" and references to numbers or groups in multiple contexts

No Comments

No Known Activations