INDEX

Explanations

occurrences of the letter 's' in various contexts, particularly in a narrative about time periods or historical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

id

-0.19

er

-0.18

-0.15

fully

-0.15

 ìķŀ

-0.15

ER

-0.15

-0.14

pedia

-0.14

POSITIVE LOGITS

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.30

-era

0.28

era

0.24

era

0.23

0.19

 dÃ¶nemde

0.18

’

0.18

decade

0.17

orgot

0.16

/'

0.16

Activations Density 0.027%

occurrences of the letter 's' in various contexts, particularly in a narrative about time periods or historical references

No Comments

No Known Activations