INDEX

Explanations

sources for additional information or further reading

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a desire for additional information

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

xtap

-0.93

idated

-0.79

ħĭ

-0.76

chwitz

-0.76

Ħ¢

-0.75

Cro

-0.75

ovie

-0.75

ļé

-0.75

Ĥ

-0.75

ĺħ

-0.73

POSITIVE LOGITS

 than

1.13

 details

0.82

 info

0.82

than

0.79

 interesting

0.77

 detailed

0.74

 VIDEOS

0.73

 options

0.73

ado

0.73

 frequently

0.71

Activations Density 0.030%

sources for additional information or further reading

phrases indicating a desire for additional information

No Comments

No Known Activations