INDEX

Explanations

references to trees in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

iras

-0.19

aries

-0.18

Î·ÏĤ

-0.17

undan

-0.16

ordes

-0.15

sten

-0.15

GGLE

-0.15

angered

-0.14

marine

-0.14

/gtest

-0.14

POSITIVE LOGITS

 trunk

0.26

house

0.23

cko

0.23

 branch

0.21

 branches

0.21

-lined

0.21

/tree

0.20

branch

0.20

 stump

0.20

(tree

0.20

Activations Density 0.024%

references to trees in various contexts

No Comments

No Known Activations