INDEX

Explanations

product or service descriptions with a focus on functionality or benefits

oai_token-act-pair · gpt-3.5-turbo

references to potential and capability in relation to products and services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 absentee

-0.74

 Torn

-0.64

gray

-0.64

 stricken

-0.59

leted

-0.57

 wrath

-0.57

 lapse

-0.57

 twilight

-0.55

 blame

-0.55

 forc

-0.55

POSITIVE LOGITS

 solves

1.51

 fulf

1.49

 embodies

1.43

 satisfies

1.40

 deserves

1.37

 achieves

1.33

 offers

1.32

 fits

1.30

 merits

1.27

 represents

1.26

Activations Density 0.382%

product or service descriptions with a focus on functionality or benefits

references to potential and capability in relation to products and services

No Comments

No Known Activations