INDEX

Explanations

references to specific quantities or measurements

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

ergy

-0.18

inia

-0.15

enia

-0.14

?p

-0.14

 occasions

-0.14

 sucker

-0.14

ITA

-0.13

aaS

-0.13

_timestamp

-0.13

iplina

-0.13

POSITIVE LOGITS

bote

0.16

niÄį

0.15

exas

0.15

wahl

0.14

setIcon

0.14

 Redistributions

0.13

icol

0.13

slack

0.13

lio

0.13

lix

0.13

Activations Density 0.275%

references to specific quantities or measurements

No Comments

No Known Activations