INDEX

Explanations

phrases or words indicating extremity or exceeding limits

oai_token-act-pair · gpt-3.5-turbo

phrases indicating limits exceeded or conditions that are no longer manageable

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pmwiki

-0.74

nian

-0.71

ICAN

-0.71

 Assass

-0.68

Ped

-0.67

ãĤ·ãĥ£

-0.66

 WATCHED

-0.65

cu

-0.64

si

-0.64

RAW

-0.62

POSITIVE LOGITS

 comprehension

0.82

iche

0.81

 bounds

0.80

 reach

0.78

 Borders

0.76

 reaching

0.76

 infancy

0.75

 mere

0.74

 Belief

0.73

esis

0.72

Activations Density 0.025%

phrases or words indicating extremity or exceeding limits

phrases indicating limits exceeded or conditions that are no longer manageable

No Comments

No Known Activations