INDEX

Explanations

references to public interests and services

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

icut

-0.15

etails

-0.15

INGER

-0.14

AXB

-0.14

-range

-0.14

minster

-0.14

wdx

-0.13

irting

-0.13

estroy

-0.13

egin

-0.13

POSITIVE LOGITS

vel

0.19

iah

0.16

Ð´Ð°Ð¼

0.15

sp

0.15

ocha

0.15

 Decomp

0.14

dea

0.14

ableView

0.14

alo

0.14

anda

0.13

Activations Density 0.032%

references to public interests and services

No Comments

No Known Activations