INDEX

Explanations

phrases related to legal processes and decisions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "was" and its variations in the context of legal or administrative proceedings

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

 associations

-0.74

ngth

-0.64

 sensibilities

-0.63

 Associ

-0.63

 entrants

-0.62

 Slate

-0.60

lighting

-0.59

Imagine

-0.59

 metaphors

-0.58

ivan

-0.57

POSITIVE LOGITS

 rescind

1.50

 cancelled

1.42

 revoked

1.40

 canceled

1.40

 withdrawn

1.30

 upheld

1.22

 overturned

1.19

 rejected

1.17

vo

1.16

 postponed

1.14

Activations Density 0.178%

phrases related to legal processes and decisions

occurrences of the word "was" and its variations in the context of legal or administrative proceedings

No Comments

No Known Activations