INDEX

Explanations

ingredients and cooking instructions in the form of quantities, spices, and herbs

oai_token-act-pair · gpt-3.5-turbo

measurements and quantities for ingredients in recipes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 evidenced

-0.80

 experien

-0.75

 manifests

-0.73

aurus

-0.73

 fielded

-0.72

arios

-0.72

wat

-0.70

 answ

-0.70

 veter

-0.69

ĸļ

-0.68

POSITIVE LOGITS

Ingredients

1.06

Optional

0.98

 Instructions

0.89

Tools

0.81

Instruct

0.81

ÂłÂłÂłÂł

0.80

protein

0.78

Salt

0.77

Â½

0.77

ulsion

0.76

Activations Density 0.091%

ingredients and cooking instructions in the form of quantities, spices, and herbs

measurements and quantities for ingredients in recipes

No Comments

No Known Activations