INDEX

Explanations

references to scientific theories and evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Kapoor

-0.17

nee

-0.16

linger

-0.14

èĦ

-0.14

erro

-0.13

quip

-0.13

iesz

-0.13

 following

-0.13

andi

-0.13

orget

-0.13

POSITIVE LOGITS

specialchars

0.15

reich

0.15

aptive

0.14

.matcher

0.14

ÑĦÐ¾ÑĢÑĤ

0.14

inerary

0.14

/TT

0.14

ypy

0.14

scape

0.13

 addCriterion

0.13

Activations Density 0.040%

references to scientific theories and evidence

No Comments

No Known Activations