INDEX

Explanations

forms of the verb "be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ppe

-0.15

 Rosenberg

-0.15

ÐµÐ¼Ð¾Ð½

-0.15

abox

-0.15

hattan

-0.15

ichel

-0.14

ston

-0.14

<path

-0.14

=__

-0.14

aisal

-0.13

POSITIVE LOGITS

fuse

0.16

alive

0.16

_fault

0.15

enza

0.14

Fuse

0.14

MDB

0.14

jam

0.14

 alive

0.14

OPSIS

0.14

tas

0.14

Activations Density 0.010%

forms of the verb "be."

No Comments

No Known Activations