INDEX

Explanations

references to academic publications and journal citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

enko

-0.16

iken

-0.15

uja

-0.15

uos

-0.14

indow

-0.14

 tran

-0.14

Ð»ÑıÐ½

-0.14

ilty

-0.14

aigned

-0.14

agan

-0.14

POSITIVE LOGITS

 letters

0.19

_letters

0.18

 Laud

0.17

letters

0.17

 letter

0.17

 LETTER

0.16

 Letters

0.16

-letter

0.16

YST

0.15

.gov

0.15

Activations Density 0.010%

references to academic publications and journal citations

No Comments

No Known Activations