INDEX

Explanations

references to claims or assertions that involve factual information or the nature of evidence in argumentation

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ä¸ĢåĪĩ

-0.14

hled

-0.13

/ar

-0.12

äºĽ

-0.12

StrictEqual

-0.12

UILTIN

-0.12

_acl

-0.12

 ragaz

-0.11

outil

-0.11

ERSHEY

-0.11

POSITIVE LOGITS

0.81

an

0.68

ä¸Ģä¸ª

0.49

 sebuah

0.40

ä¸ĢåĢĭ

0.39

0.38

æĺ¯ä¸Ģä¸ª

0.36

,a

0.35

—a

0.34

Activations Density 7.032%

references to claims or assertions that involve factual information or the nature of evidence in argumentation

No Known Activations