INDEX

Explanations

words related to actions or activities, particularly those involving planning, decision-making, or structured processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ijd

-0.15

avan

-0.14

vit

-0.14

cape

-0.14

ktop

-0.14

ael

-0.14

usan

-0.14

Î´Î¹Î¿

-0.13

avern

-0.13

aul

-0.13

POSITIVE LOGITS

 Kang

0.18

etes

0.15

ENTIAL

0.15

 Meadows

0.15

jeta

0.15

ungs

0.15

UNG

0.15

swire

0.14

roys

0.14

angelo

0.14

Activations Density 0.014%

words related to actions or activities, particularly those involving planning, decision-making, or structured processes

No Comments

No Known Activations