INDEX

Explanations

phrases related to an on-going process or activity

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a process or ongoing action

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

ï¸ı

-0.69

 Bust

-0.68

Flags

-0.67

Champ

-0.66

Edge

-0.65

HD

-0.64

oons

-0.63

oft

-0.62

orah

-0.61

 Flags

-0.61

POSITIVE LOGITS

 obtaining

0.89

 progress

0.87

 arranging

0.83

 evaluating

0.83

 transferring

0.80

 applying

0.79

 extracting

0.79

 resolving

0.78

 incarcer

0.77

 constructing

0.77

Activations Density 0.105%

phrases related to an on-going process or activity

phrases that indicate a process or ongoing action

No Comments

No Known Activations