INDEX

Explanations

references to foresight or thinking ahead

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

atab

-0.17

uchen

-0.16

iten

-0.15

legalArgumentException

-0.14

ulu

-0.14

ÎŃ

-0.14

 sublicense

-0.14

Sax

-0.14

isoft

-0.14

edik

-0.14

POSITIVE LOGITS

alls

0.15

={['

0.14

olg

0.14

lock

0.14

ball

0.13

ui

0.13

umes

0.13

.reducer

0.13

izons

0.13

Pres

0.13

Activations Density 0.003%

references to foresight or thinking ahead

No Comments

No Known Activations