INDEX

Explanations

mentions of benefits or advantages of various actions or technologies

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "benefits" in various contexts

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

rez

-0.67

opic

-0.66

ãĥ£

-0.63

zar

-0.63

ger

-0.62

 Penal

-0.61

ones

-0.61

ini

-0.60

dimension

-0.60

FP

-0.60

POSITIVE LOGITS

benefit

1.07

 conferred

1.05

 outwe

1.01

 outweigh

0.99

 benefit

0.89

 afforded

0.85

 benefits

0.85

Benef

0.80

ocratic

0.79

effects

0.79

Activations Density 0.021%

mentions of benefits or advantages of various actions or technologies

mentions of the term "benefits" in various contexts

No Comments

No Known Activations