INDEX

Explanations

significant events or impactful actions

oai_token-act-pair · gpt-3.5-turbo

terms associated with performance or outcomes related to events or incidents

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

 fraction

-0.57

FN

-0.56

atorium

-0.55

IFE

-0.53

ife

-0.52

nick

-0.52

 Canal

-0.51

pse

-0.50

ieth

-0.48

reci

-0.48

POSITIVE LOGITS

poons

1.16

hips

1.14

uits

0.97

paces

0.91

 ranging

0.91

heet

0.90

hip

0.88

ets

0.85

ettings

0.85

mith

0.85

Activations Density 0.456%

significant events or impactful actions

terms associated with performance or outcomes related to events or incidents

No Comments

No Known Activations