INDEX

Explanations

terms that are important or applicable to the context being discussed

oai_token-act-pair · gpt-3.5-turbo

the term "relevant" and its applicability in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

MAX

-0.73

illon

-0.71

maid

-0.68

ache

-0.66

alone

-0.65

 quadru

-0.63

 Gets

-0.63

 waits

-0.62

 vows

-0.61

issors

-0.60

POSITIVE LOGITS

 relevant

3.63

 pertinent

2.69

relevant

2.48

 applicable

2.05

 appropriate

1.87

 relevance

1.80

levant

1.77

 irrelevant

1.66

 salient

1.56

 meaningful

1.39

Activations Density 0.007%

terms that are important or applicable to the context being discussed

the term "relevant" and its applicability in various contexts

No Comments

No Known Activations