INDEX

Explanations

mentions of urging or advising

oai_token-act-pair · gpt-3.5-turbo

instances of the word "urge" and its variations 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fucking

-0.68

 match

-0.68

thing

-0.66

 fuck

-0.61

 guys

-0.61

Plex

-0.60

 matching

-0.60

ater

-0.59

 matches

-0.59

 hindsight

-0.58

POSITIVE LOGITS

 urged

3.13

 cautioned

2.02

 advised

2.01

 urges

1.97

 warned

1.93

 encouraged

1.92

 urging

1.72

 congratulated

1.66

 urge

1.61

 advocated

1.60

Activations Density 0.017%

mentions of urging or advising

instances of the word "urge" and its variations in various contexts

No Comments

No Known Activations