INDEX

Explanations

instances of communication or directives in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

CTL

-0.14

tier

-0.14

oples

-0.14

setDefault

-0.14

 ÑĸÐ½ÑĤÐµÑĢÐµÑģ

-0.13

ahan

-0.13

acs

-0.13

 Fulton

-0.13

.ct

-0.13

shal

-0.13

POSITIVE LOGITS

esson

0.17

AEA

0.16

ght

0.15

odka

0.15

ayscale

0.15

-fw

0.15

 incoming

0.15

incoming

0.15

 Wein

0.14

Activations Density 0.103%

instances of communication or directives in the text

No Comments

No Known Activations