INDEX

Explanations

indicators of connections or references to related topics

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

apse

-0.18

ippet

-0.16

ramework

-0.16

vala

-0.15

Ð¾Ð±Ð°

-0.15

ekim

-0.15

usto

-0.14

iales

-0.14

Vu

-0.14

iais

-0.14

POSITIVE LOGITS

 Garr

0.16

 Gall

0.16

 Items

0.15

_DIRS

0.15

rian

0.14

cape

0.14

Ish

0.14

Ã¶h

0.14

.GetItem

0.14

ockets

0.14

Activations Density 0.008%

indicators of connections or references to related topics

No Comments

No Known Activations