INDEX

Explanations

sequences or patterns within text or data

oai_token-act-pair · gpt-3.5-turbo

mentions of sequences, particularly in a mathematical or procedural context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ryn

-0.85

aghd

-0.80

sburgh

-0.80

sie

-0.78

lying

-0.77

lisher

-0.76

nai

-0.72

young

-0.71

hoff

-0.70

isle

-0.70

POSITIVE LOGITS

 Sequence

1.00

 sequence

0.88

 sequences

0.84

ĸļ

0.83

uality

0.74

 unfold

0.73

 repeats

0.73

alysis

0.70

 sequencing

0.64

 weaving

0.64

Activations Density 0.028%

sequences or patterns within text or data

mentions of sequences, particularly in a mathematical or procedural context

No Comments

No Known Activations