INDEX

Explanations

terms or phrases related to leaks, especially in the context of information or substances escaping from containment

oai_token-act-pair · gpt-3.5-turbo

words related to leaks or leakage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

iterranean

-0.83

zees

-0.72

issors

-0.71

aldi

-0.70

oran

-0.66

eele

-0.65

============

-0.64

HCR

-0.63

uliffe

-0.63

=-=-=-=-

-0.63

POSITIVE LOGITS

 leak

0.90

 leaks

0.88

 leaking

0.84

inventoryQuantity

0.82

acious

0.81

Leaks

0.81

0.75

 leakage

0.70

ily

0.69

 oily

0.67

Activations Density 0.012%

terms or phrases related to leaks, especially in the context of information or substances escaping from containment

words related to leaks or leakage

No Comments

No Known Activations