INDEX

Explanations

triggering events or incidents that lead to significant reactions or consequences

oai_token-act-pair · gpt-3.5-turbo

instances of the word "sparked" indicating significant events or reactions

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

esan

-0.76

uters

-0.66

endi

-0.66

 Neighbor

-0.64

ians

-0.60

ibble

-0.60

itte

-0.60

 route

-0.58

 wives

-0.57

iolet

-0.54

POSITIVE LOGITS

TextColor

0.92

rings

0.91

actionDate

0.89

 pandemonium

0.79

interstitial

0.76

rament

0.75

 sparking

0.75

 stir

0.75

 anew

0.73

 sparked

0.73

Activations Density 0.047%

triggering events or incidents that lead to significant reactions or consequences

instances of the word "sparked" indicating significant events or reactions

No Comments

No Known Activations