INDEX

Explanations

references to government agencies and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ces

-0.15

 whom

-0.14

preh

-0.14

terraform

-0.14

/functions

-0.13

que

-0.13

ook

-0.13

apesh

-0.13

kovi

-0.13

lus

-0.13

POSITIVE LOGITS

 itself

0.17

's

0.17

ARING

0.16

wide

0.16

-community

0.15

Wyn

0.14

avin

0.14

inki

0.14

istrar

0.14

atisch

0.14

Activations Density 0.137%

references to government agencies and institutions

No Comments

No Known Activations