INDEX

Explanations

references to political figures and their characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

eka

-0.16

 Ambient

-0.15

kontakte

-0.15

.docs

-0.15

èĪį

-0.15

zier

-0.15

mploy

-0.15

ertainment

-0.14

intendent

-0.14

UNCH

-0.14

POSITIVE LOGITS

 moderate

0.23

 comb

0.20

 establishment

0.20

 Establishment

0.20

 figure

0.20

 cent

0.19

 icon

0.19

 prote

0.19

 cerebral

0.19

 consum

0.19

Activations Density 0.121%

references to political figures and their characteristics

No Comments

No Known Activations