INDEX

Explanations

quantitative expressions indicating abundance and significance

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ibt

-0.15

ennen

-0.15

æ®

-0.15

_exist

-0.14

anded

-0.14

av

-0.14

ula

-0.13

 harb

-0.13

 minded

-0.13

with

-0.13

POSITIVE LOGITS

 happening

0.24

 else

0.19

Else

0.18

 wrong

0.17

 riding

0.17

 Else

0.17

åıĳçĶŁ

0.17

 room

0.16

inging

0.16

 overlap

0.16

Activations Density 0.048%

quantitative expressions indicating abundance and significance

No Comments

No Known Activations