INDEX

Explanations

the preposition "in" across 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 @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

acz

-0.15

oux

-0.15

oved

-0.15

 cass

-0.15

ollen

-0.15

uzey

-0.14

ocz

-0.14

NOWLED

-0.14

chet

-0.14

formed

-0.14

POSITIVE LOGITS

 SolidColorBrush

0.15

atar

0.15

istani

0.15

Circular

0.15

 Circular

0.14

tit

0.14

mast

0.14

ndata

0.14

Stick

0.14

IMIT

0.14

Activations Density 0.015%

the preposition "in" across various contexts

No Comments

No Known Activations