INDEX

Explanations

details related to the context and nuances of various subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

aul

-0.14

addtogroup

-0.14

elib

-0.14

alnum

-0.13

ÏĨÎ±

-0.13

ëĤĺëĿ¼

-0.13

akin

-0.13

abad

-0.13

 Photon

-0.12

?=.*

-0.12

POSITIVE LOGITS

 context

0.69

 background

0.64

context

0.58

 Context

0.54

 Background

0.50

-context

0.50

background

0.49

èĥĮæĻ¯

0.49

 CONTEXT

0.48

_context

0.48

Activations Density 0.127%

details related to the context and nuances of various subjects

No Comments

No Known Activations