INDEX

Explanations

terms related to quests or missions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "quest" and its variations

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

Effective

-0.80

ews

-0.71

ACP

-0.71

Charlie

-0.70

 Hilton

-0.70

 Charlie

-0.66

 Cohen

-0.65

ENC

-0.64

oli

-0.64

 Peel

-0.63

POSITIVE LOGITS

 quest

3.64

 quests

2.52

 Quest

1.94

Quest

1.72

quest

1.54

 pursuit

1.52

 mission

1.41

 crusade

1.36

 hunt

1.32

 search

1.29

Activations Density 0.016%

terms related to quests or missions

occurrences of the word "quest" and its variations

No Comments

No Known Activations