INDEX

Explanations

references to nuclear events and their consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ransition

-0.17

illez

-0.16

egers

-0.16

jadi

-0.16

eldom

-0.15

è

-0.14

icht

-0.14

atus

-0.14

aggio

-0.14

POSITIVE LOGITS

tw

0.18

æŀ

0.16

afen

0.15

IRR

0.15

 Matt

0.15

 deser

0.14

¬Ĥ

0.14

Bes

0.14

otta

0.14

otton

0.14

Activations Density 0.232%

references to nuclear events and their consequences

No Comments

No Known Activations