INDEX

Explanations

modal verbs that indicate possibility or capability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ience

-0.16

 Ø§ÙĦØ¸

-0.14

.isSuccessful

-0.14

èĢĥ

-0.14

obe

-0.14

Ð»Ð¸ÐºÐ°

-0.14

ance

-0.14

eping

-0.13

ru

-0.13

ailing

-0.13

POSITIVE LOGITS

ãĥ¼ãĥª

0.16

ysqli

0.16

aar

0.16

lesi

0.15

ROTO

0.14

ukes

0.14

ÃŃcia

0.14

.echo

0.14

ihar

0.14

loh

0.14

Activations Density 0.070%

modal verbs that indicate possibility or capability

No Comments

No Known Activations