INDEX

Explanations

actions or instructions related to cooking or preparing food

oai_token-act-pair · gpt-3.5-turbo

phrases indicating location or inclusion within a context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iffs

-0.63

rn

-0.62

Lew

-0.58

fn

-0.58

-0.56

 defect

-0.56

iae

-0.56

itton

-0.53

 Benn

-0.53

orem

-0.52

POSITIVE LOGITS

 fridge

1.08

 blender

1.07

 salads

1.07

 bowls

1.05

 jars

1.02

redients

1.02

 refrigerator

1.02

 batches

0.96

 microwave

0.95

 boiling

0.93

Activations Density 0.097%

actions or instructions related to cooking or preparing food

phrases indicating location or inclusion within a context

No Comments

No Known Activations