INDEX

Explanations

numerical data and structured information tables

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

 Uncomment

-0.16

egin

-0.16

aign

-0.16

enable

-0.15

hoa

-0.15

owie

-0.15

def

-0.14

eg

-0.14

eg

-0.14

 mechanism

-0.13

POSITIVE LOGITS

 Loved

0.15

#ad

0.15

ÑģÑĮÐ¾Ð³Ð¾

0.14

nete

0.14

nnen

0.14

uggy

0.14

upo

0.14

£¼

0.14

_FRAMEBUFFER

0.14

utterstock

0.13

Activations Density 0.031%

numerical data and structured information tables

No Comments

No Known Activations