INDEX

Explanations

repetitive phrases or ideas throughout the text

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

ãģĤãĤĭ

-0.16

esta

-0.15

xs

-0.15

Tul

-0.14

 Contr

-0.14

å¦Ĥä¸ĭ

-0.14

ised

-0.14

op

-0.13

ein

-0.13

own

-0.13

POSITIVE LOGITS

/th

0.24

 particular

0.21

iner

0.18

chy

0.17

ched

0.17

curity

0.16

Dll

0.16

 Ð¶Ðµ

0.15

zelf

0.15

 same

0.15

Activations Density 0.147%

repetitive phrases or ideas throughout the text

No Comments

No Known Activations