INDEX

Explanations

imperative verbs and phrases indicating requests or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ssa

-0.15

itorio

-0.15

Ĩ

-0.15

ijn

-0.14

igure

-0.14

stvo

-0.14

ificaciones

-0.14

esson

-0.14

	fflush

-0.13

setattr

-0.13

POSITIVE LOGITS

lias

0.16

kuk

0.15

ctor

0.14

zee

0.14

ILA

0.14

allas

0.14

ople

0.14

Gig

0.14

iams

0.14

cala

0.14

Activations Density 0.030%

imperative verbs and phrases indicating requests or actions

No Comments

No Known Activations