INDEX

Explanations

instances of specific names and places related to events or historical references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 @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

à¸£à¸ģ

-0.16

 bergen

-0.15

 Yorker

-0.15

 prelim

-0.15

afen

-0.15

Ordinal

-0.14

Likes

-0.14

èĶ

-0.14

pedia

-0.14

 superClass

-0.14

POSITIVE LOGITS

 absol

0.25

ar

0.23

 stud

0.22

 absolute

0.19

bes

0.19

 prom

0.18

 pend

0.17

 heir

0.17

bet

0.16

ilha

0.16

Activations Density 0.022%

instances of specific names and places related to events or historical references

No Comments

No Known Activations