INDEX

Explanations

references to cannons and artillery-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Ãªt

-0.17

olon

-0.16

olie

-0.15

PLICATION

-0.15

oppers

-0.14

.dep

-0.14

Ã³mo

-0.14

gal

-0.13

ded

-0.13

.getBlock

-0.13

POSITIVE LOGITS

ichael

0.16

arium

0.14

 Sterling

0.14

aurant

0.14

vale

0.14

ajÃŃ

0.14

vere

0.14

feld

0.14

aires

0.14

ysis

0.14

Activations Density 0.010%

references to cannons and artillery-related terms

No Comments

No Known Activations