INDEX

Explanations

proper names and affiliations in academic or research contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

urance

-0.18

CallCheck

-0.18

bih

-0.17

urus

-0.16

alars

-0.15

elian

-0.15

ucken

-0.15

OLON

-0.15

aleigh

-0.15

anuts

-0.15

POSITIVE LOGITS

 Hast

0.17

 Machine

0.17

im

0.16

Prov

0.15

 stat

0.15

-stat

0.15

pac

0.15

_stat

0.15

 machine

0.15

ban

0.14

Activations Density 0.074%

proper names and affiliations in academic or research contexts

No Comments

No Known Activations