INDEX

Explanations

elements and phrases related to legal and procedural concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ikan

-0.14

Scope

-0.14

idak

-0.14

ubl

-0.14

 miejsc

-0.14

 nIndex

-0.14

ult

-0.13

 ÙĤÙĬ

-0.13

ream

-0.13

imonial

-0.13

POSITIVE LOGITS

edly

0.17

avs

0.14

utar

0.14

ahoo

0.14

Tau

0.14

avigate

0.13

Ø¨Ø±ÛĮ

0.13

aves

0.13

smith

0.13

elt

0.13

Activations Density 0.339%

elements and phrases related to legal and procedural concepts

No Comments

No Known Activations