INDEX

Explanations

phrases indicating complexity or intensity in experiences or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

imos

-0.17

ekim

-0.16

pread

-0.16

 ãĢĳ

-0.15

inders

-0.15

Dao

-0.15

ÎķÎł

-0.14

ÏĦÎ¹Îº

-0.14

à¹ĭ

-0.14

 Bair

-0.14

POSITIVE LOGITS

hw

0.19

UTTON

0.15

leta

0.15

 sorte

0.14

 sort

0.14

-NLS

0.14

lope

0.14

ys

0.14

ioni

0.14

imson

0.13

Activations Density 0.018%

phrases indicating complexity or intensity in experiences or narratives

No Comments

No Known Activations