INDEX

Explanations

references to tables and statistical data in documents

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atile

-0.17

iculty

-0.15

artment

-0.14

ering

-0.14

emen

-0.14

 Merc

-0.13

 millennium

-0.13

_matched

-0.13

rames

-0.13

elling

-0.13

POSITIVE LOGITS

oya

0.22

ÙĪØ§Ø±Ùĩ

0.17

aldi

0.17

tabla

0.17

 table

0.17

AdapterManager

0.17

/tab

0.17

(table

0.16

aux

0.16

au

0.16

Activations Density 0.026%

references to tables and statistical data in documents

No Comments

No Known Activations