INDEX

Explanations

terms related to mathematical parameters and their relationships in models

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

isini

-0.17

obo

-0.15

HasBeenSet

-0.15

otten

-0.15

///<

-0.15

oz

-0.14

Datum

-0.14

ozem

-0.14

usty

-0.14

neau

-0.14

POSITIVE LOGITS

 Gret

0.17

rang

0.14

 depress

0.14

 Recovery

0.14

 strokes

0.14

Ã¤tz

0.14

aec

0.14

jsc

0.14

eway

0.13

quip

0.13

Activations Density 0.185%

terms related to mathematical parameters and their relationships in models

No Comments

No Known Activations