INDEX

Explanations

phrases indicating probability or likelihood pertaining to events or outcomes

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

.au

-0.16

Ã£ng

-0.16

pe

-0.15

ulary

-0.15

cca

-0.14

ular

-0.14

jerne

-0.14

ivas

-0.13

chw

-0.13

chie

-0.13

POSITIVE LOGITS

hood

0.41

 hood

0.29

 candidates

0.22

 soon

0.20

;y

0.20

ilty

0.19

candidates

0.18

 Hood

0.18

 candidate

0.17

 Candidates

0.17

Activations Density 0.029%

phrases indicating probability or likelihood pertaining to events or outcomes

No Comments

No Known Activations