INDEX

Explanations

terminology and concepts related to formal processes and logic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 redund

-0.15

Ð²Ð°Ð¼Ð¸

-0.14

.echo

-0.14

bol

-0.14

hek

-0.13

onz

-0.13

quist

-0.13

 octave

-0.13

tan

-0.13

 bolts

-0.13

POSITIVE LOGITS

 trace

0.24

 traces

0.22

 Trace

0.21

 valuation

0.21

trace

0.19

_trace

0.19

(trace

0.18

Trace

0.18

 stutter

0.17

 fresh

0.17

Activations Density 0.025%

terminology and concepts related to formal processes and logic

No Comments

No Known Activations