INDEX

Explanations

text related to commands or instructions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ll" as a contraction or abbreviation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

EStream

-0.83

 guiActiveUn

-0.79

 behavi

-0.75

ãĥĩ

-0.73

uliffe

-0.71

lished

-0.70

*/(

-0.69

ItemTracker

-0.66

 slack

-0.65

¥ŀ

-0.64

POSITIVE LOGITS

oyd

1.30

ounge

1.09

inois

1.09

ength

1.03

uminati

1.02

uci

0.98

ll

0.98

ows

0.97

sburgh

0.95

ices

0.89

Activations Density 0.031%

text related to commands or instructions

occurrences of the word "ll" as a contraction or abbreviation

No Comments

No Known Activations