INDEX

Explanations

mentions of someone or something being 'chosen'

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "chosen"

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uther

-0.72

Net

-0.71

emic

-0.70

urst

-0.69

ear

-0.67

-0.66

pat

-0.66

auer

-0.66

POSITIVE LOGITS

 chosen

1.13

 randomly

0.88

 choosing

0.84

 chose

0.78

 arbitrarily

0.77

 chooses

0.77

 choice

0.76

ĨĴ

0.73

selection

0.72

lists

0.70

Activations Density 0.007%

mentions of someone or something being 'chosen'

occurrences of the word "chosen"

No Comments

No Known Activations