INDEX

Explanations

intensifiers and adverbs of degree that enhance the meaning of adjectives and verbs

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

iesz

-0.19

.scalablytyped

-0.16

ocity

-0.15

indow

-0.15

 sanit

-0.14

áº©n

-0.14

 Giles

-0.14

_INITIALIZER

-0.13

ependency

-0.13

HORT

-0.13

POSITIVE LOGITS

-ton

0.15

ande

0.15

 propos

0.15

shaw

0.15

-tone

0.15

anch

0.14

 Tone

0.14

tor

0.14

Å¡ak

0.14

oin

0.13

Activations Density 0.130%

intensifiers and adverbs of degree that enhance the meaning of adjectives and verbs

No Comments

No Known Activations