INDEX

Explanations

words indicative of social and political controversies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

OF

-0.16

brero

-0.15

uellement

-0.15

.LayoutStyle

-0.15

onnen

-0.14

ovacÃŃ

-0.14

icularly

-0.14

ansen

-0.13

ÑĤÑĶ

-0.13

Ã½n

-0.13

POSITIVE LOGITS

 surrounding

0.37

 Surround

0.33

 over

0.31

 surround

0.30

 surrounds

0.25

 about

0.22

 around

0.21

over

0.20

about

0.18

åĽ´

0.18

Activations Density 0.098%

words indicative of social and political controversies

No Comments

No Known Activations