INDEX

Explanations

references to ongoing discussions and collaborative efforts

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

oÅĪ

-0.14

Intro

-0.13

emony

-0.13

æķ¦

-0.13

iminal

-0.13

imeo

-0.12

_defs

-0.12

ÑĢÐµÐ¶

-0.12

encies

-0.12

.ps

-0.12

POSITIVE LOGITS

 discussion

0.81

 discuss

0.73

 discussing

0.71

 discussions

0.68

 Discussion

0.66

discussion

0.65

 discussed

0.63

 discusses

0.61

 Discuss

0.61

Discussion

0.60

Activations Density 0.188%

references to ongoing discussions and collaborative efforts

No Comments

No Known Activations