INDEX

Explanations

terms related to benefits, advantages, or positive outcomes

oai_token-act-pair · gpt-3.5-turbo

references to various types of benefits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ãĥ£

-0.81

²¾

-0.76

taboola

-0.67

ini

-0.66

Plot

-0.66

wcsstore

-0.62

ergy

-0.62

Else

-0.61

tone

-0.61

rez

-0.61

POSITIVE LOGITS

 benefits

1.20

benefit

1.20

 benefit

1.07

 Benefits

1.03

benef

1.03

Benef

1.00

 synerg

0.94

 conferred

0.93

 accrued

0.84

 disadvant

0.83

Activations Density 0.017%

terms related to benefits, advantages, or positive outcomes

references to various types of benefits

No Comments

No Known Activations