INDEX

Explanations

instances of significant phrases or titles related to documents or publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

.compose

-0.16

uky

-0.15

TestId

-0.15

uiten

-0.15

tab

-0.14

pun

-0.14

/pdf

-0.14

pun

-0.14

 cross

-0.14

TAB

-0.14

POSITIVE LOGITS

item

0.16

á»ĵn

0.15

com

0.15

ek

0.14

engine

0.14

usted

0.14

exe

0.14

çĮ®

0.14

 æĪ

0.14

ernity

0.14

Activations Density 0.002%

instances of significant phrases or titles related to documents or publications

No Comments

No Known Activations