INDEX

Explanations

references to operators in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

allo

-0.15

Ð»ÑıÐ½

-0.15

qa

-0.14

eners

-0.14

quer

-0.14

ega

-0.14

askan

-0.14

Å¼e

-0.14

ali

-0.14

POSITIVE LOGITS

ooth

0.15

ì²´

0.15

ehler

0.15

regunta

0.15

anzi

0.15

untu

0.14

hlen

0.14

razier

0.14

oze

0.14

olec

0.14

Activations Density 0.006%

references to operators in various contexts

No Comments

No Known Activations