INDEX

Explanations

quantitative phrases or expressions related to estimates and approximations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

á»ĳc

-0.16

rzy

-0.14

ring

-0.14

Ú¯Ø§ÙĨÛĮ

-0.14

ress

-0.14

ensch

-0.13

rk

-0.13

ı

-0.13

urn

-0.13

ynchronously

-0.13

POSITIVE LOGITS

ãĥ«ãĥĪ

0.17

lane

0.16

uncio

0.15

aise

0.15

 Dear

0.15

Ø¨ÛĮØ±

0.15

kla

0.15

Ray

0.14

MUX

0.14

 Latter

0.14

Activations Density 0.067%

quantitative phrases or expressions related to estimates and approximations

No Comments

No Known Activations