INDEX

Explanations

words related to storytelling and coherent plotlines

oai_token-act-pair · gpt-3.5-turbo

references to narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

upiter

-0.83

unte

-0.81

rador

-0.77

gans

-0.77

ateurs

-0.75

ugs

-0.74

inking

-0.72

thening

-0.72

ribute

-0.71

achev

-0.70

POSITIVE LOGITS

 narratives

1.22

 narrative

1.09

 Narr

0.94

Narr

0.93

ufact

0.85

 arcs

0.82

 storylines

0.80

 unfolds

0.76

escription

0.75

 è£ıè

0.74

Activations Density 0.010%

words related to storytelling and coherent plotlines

references to narratives

No Comments

No Known Activations