INDEX

Explanations

words related to influence, impact, and consequences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the impact of actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

atin

-0.76

etter

-0.63

 mash

-0.62

 specimen

-0.62

 confess

-0.58

 Corpus

-0.58

ft

-0.57

 solicitation

-0.55

 behold

-0.55

Cass

-0.55

POSITIVE LOGITS

 repercussions

1.34

 implications

1.31

 ramifications

1.31

 consequences

1.24

 profound

1.10

 ripple

1.06

 significance

1.04

 unintended

0.99

 effects

0.99

 relevance

0.98

Activations Density 0.161%

words related to influence, impact, and consequences

phrases indicating the impact of actions or events

No Comments

No Known Activations