INDEX

Explanations

mentions of President Joe Biden and 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 @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

aub

-0.16

vale

-0.16

nutÃŃ

-0.15

ses

-0.15

-fontawesome

-0.15

Ø±Ø¯

-0.14

occo

-0.14

egra

-0.14

OURCES

-0.14

inyin

-0.14

POSITIVE LOGITS

ior

0.20

ium

0.18

0.16

iores

0.14

frey

0.14

 Washing

0.14

0.13

bean

0.13

omics

0.13

Uns

0.13

Activations Density 0.007%

mentions of President Joe Biden and related terms

No Comments

No Known Activations