INDEX

Explanations

numbers in a specific format

oai_token-act-pair · gpt-3.5-turbo

repeated patterns of punctuation or numerical sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uggest

-0.65

ngth

-0.65

 melanch

-0.64

 fantasies

-0.58

 reconc

-0.58

gad

-0.58

vag

-0.55

 Cros

-0.55

ioned

-0.55

ashtra

-0.54

POSITIVE LOGITS

1.13

0.98

0.97

0.93

0.91

0.90

0.89

0.88

0.86

Activations Density 0.110%

numbers in a specific format

repeated patterns of punctuation or numerical sequences

No Comments

No Known Activations