INDEX

Explanations

coding structures and commands related to data manipulation and graphical representation

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

-0.15

emmel

-0.14

uttgart

-0.14

etten

-0.13

fea

-0.13

 ncols

-0.13

hal

-0.13

ffa

-0.12

Nd

-0.12

 kitt

-0.12

POSITIVE LOGITS

 GANG

0.15

/errors

0.14

sik

0.14

 series

0.13

ENTA

0.13

WidgetItem

0.13

ITA

0.13

 Debt

0.13

Dash

0.13

tod

0.13

Activations Density 0.045%

coding structures and commands related to data manipulation and graphical representation

No Comments

No Known Activations