INDEX

Explanations

words related to complexity or complicated situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

pr

-0.17

psilon

-0.17

COM

-0.17

Ã¤ge

-0.16

prs

-0.16

thood

-0.15

latex

-0.15

labs

-0.15

amoto

-0.15

eps

-0.15

POSITIVE LOGITS

 Comp

0.23

ensation

0.23

 comp

0.23

.comp

0.23

(comp

0.21

ounding

0.21

licated

0.20

arrison

0.20

aign

0.20

atri

0.20

Activations Density 0.010%

words related to complexity or complicated situations

No Comments

No Known Activations