INDEX

Explanations

text related to specific designations or classifications

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "designated" and its variations, indicating a focus on formal classifications or assignments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

restling

-0.73

src

-0.71

Ev

-0.69

ev

-0.68

lip

-0.68

puff

-0.66

sea

-0.61

abuse

-0.60

escape

-0.60

answer

-0.59

POSITIVE LOGITS

 designated

3.52

 designate

2.12

 designation

1.63

 assigned

1.56

 allotted

1.48

 authorized

1.44

 selected

1.41

 labeled

1.39

 specified

1.39

 chosen

1.39

Activations Density 0.009%

text related to specific designations or classifications

occurrences of the word "designated" and its variations, indicating a focus on formal classifications or assignments

No Comments

No Known Activations