INDEX

Explanations

specific symbols or letters, possibly indicating special formatting or coding contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

Å¯v

-0.15

arend

-0.15

egis

-0.15

apol

-0.14

ward

-0.14

inox

-0.14

orne

-0.14

ospace

-0.14

correct

-0.13

enced

-0.13

POSITIVE LOGITS

 Israeli

0.23

 Palestinian

0.21

 Beit

0.20

 Palestinians

0.20

 Palestine

0.19

 Israel

0.19

Israeli

0.18

 Palestin

0.17

 ÙģÙĦØ³

0.17

Pale

0.17

Activations Density 0.002%

specific symbols or letters, possibly indicating special formatting or coding contexts

No Comments

No Known Activations