INDEX

Explanations

references to causes and effects, particularly in relation to problems or events

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

ti

-0.17

ize

-0.17

 caus

-0.16

 causal

-0.16

ird

-0.16

news

-0.16

posable

-0.15

izable

-0.15

itr

-0.15

 cause

-0.15

POSITIVE LOGITS

-effect

0.31

 cÃ©l

0.27

 cele

0.26

effect

0.20

ways

0.18

way

0.17

 celebr

0.17

Effect

0.17

lessly

0.17

lesh

0.17

Activations Density 0.024%

references to causes and effects, particularly in relation to problems or events

No Comments

No Known Activations