INDEX

Explanations

phrases indicating a specific sequence of steps or instructions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "The" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

SHARE

-0.77

-0.74

LGBT

-0.72

Brexit

-0.71

Scotland

-0.68

abuse

-0.67

Pol

-0.66

SPONSORED

-0.66

ghazi

-0.66

AIDS

-0.66

POSITIVE LOGITS

oret

1.52

 easiest

1.47

 downside

1.36

 simplest

1.31

 drawback

1.30

 resultant

1.22

 resulting

1.18

 difference

1.18

 quickest

1.17

 goal

1.17

Activations Density 0.346%

phrases indicating a specific sequence of steps or instructions

occurrences of the word "The" across various contexts

No Comments

No Known Activations