INDEX

Explanations

references to desks or workspace areas

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

iage

-0.16

chest

-0.15

heck

-0.14

sing

-0.14

issant

-0.14

 Regents

-0.14

zsche

-0.14

 Pediatric

-0.13

oren

-0.13

POSITIVE LOGITS

æĻ¶

0.15

 Khal

0.14

urdy

0.14

ÐµÐ½Ðµ

0.14

_hostname

0.14

Comparator

0.14

ene

0.13

Î¯Î½Î±

0.13

.','

0.13

artment

0.13

Activations Density 0.005%

references to desks or workspace areas

No Comments

No Known Activations