INDEX

Explanations

terms related to legal professionals and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

linkplain

-0.15

à¹īà¸§

-0.15

PRI

-0.15

 ÙĪØ§Ø³

-0.15

esin

-0.14

sume

-0.14

ighthouse

-0.14

 MetroFramework

-0.13

llx

-0.13

 Ð¿Ð¾Ð´Ð°Ð²

-0.13

POSITIVE LOGITS

Yug

0.16

tsy

0.16

Å¥

0.16

 Chore

0.15

ayah

0.14

oes

0.14

AMS

0.14

uta

0.14

chia

0.13

.RowCount

0.13

Activations Density 0.007%

terms related to legal professionals and processes

No Comments

No Known Activations