INDEX

Explanations

instances of formatting or structural elements in a document, like headings or figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ucci

-0.15

ÑĢÐµÐ»

-0.15

leting

-0.15

isin

-0.15

grim

-0.14

è¦ª

-0.14

itters

-0.14

esco

-0.14

zman

-0.14

AFE

-0.13

POSITIVE LOGITS

odor

0.15

 CHARSET

0.15

-END

0.15

-addons

0.14

 serpent

0.14

LLL

0.14

CAS

0.14

çĪ½

0.13

GetWidth

0.13

.instrument

0.13

Activations Density 0.005%

instances of formatting or structural elements in a document, like headings or figures

No Comments

No Known Activations