INDEX

Explanations

legal or administrative terms related to reviewing and evaluating documents or decisions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "review" and related terms in a legal or evaluative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Alias

-0.71

enne

-0.70

ARM

-0.67

 Chocolate

-0.65

-+

-0.65

Minecraft

-0.65

PIN

-0.64

keys

-0.64

\">

-0.63

alls

-0.62

POSITIVE LOGITS

 feasibility

1.13

 merits

0.96

 legality

0.82

 effectiveness

0.82

 favorably

0.82

 judgment

0.81

 adequ

0.80

 contents

0.79

 totality

0.78

 dossier

0.77

Activations Density 0.238%

legal or administrative terms related to reviewing and evaluating documents or decisions

instances of the word "review" and related terms in a legal or evaluative context

No Comments

No Known Activations