INDEX

Explanations

terms indicating timelines, past events, and specific references to previous situations or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

quisition

-0.15

mination

-0.15

asty

-0.14

pline

-0.14

Occurred

-0.14

activation

-0.14

fluence

-0.14

antage

-0.14

sumer

-0.13

affle

-0.13

POSITIVE LOGITS

uably

0.24

etheless

0.23

-wise

0.23

umably

0.21

sequently

0.20

 wise

0.20

 Ð¶Ðµ

0.20

ingly

0.19

prisingly

0.19

istically

0.18

Activations Density 0.308%

terms indicating timelines, past events, and specific references to previous situations or entities

No Comments

No Known Activations