INDEX

Explanations

instances of the word "after" and its variations, indicating temporal references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

keh

-0.15

ap

-0.14

ÑĢÐ°Ð¶Ð´

-0.14

umes

-0.14

kiem

-0.14

airo

-0.14

pers

-0.13

;amp

-0.13

inc

-0.13

mag

-0.13

POSITIVE LOGITS

noon

0.21

ward

0.17

alet

0.16

agua

0.16

 Ø¯ÙĪØ¨

0.14

iera

0.14

ÐĬ

0.14

Warn

0.14

-initial

0.14

istrovstvÃŃ

0.14

Activations Density 0.064%

instances of the word "after" and its variations, indicating temporal references

No Comments

No Known Activations