INDEX

Explanations

headlines or titles that contain the term "Story"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Story" in the text

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

İĭ

-0.77

 stocking

-0.72

ibilities

-0.70

»Ĵ

-0.68

ibly

-0.66

 duty

-0.66

missible

-0.65

ibles

-0.64

 retirees

-0.63

orpor

-0.63

POSITIVE LOGITS

telling

1.05

 Story

0.93

 Transcript

0.85

Maker

0.83

 Continued

0.79

 Stories

0.77

story

0.77

line

0.77

Story

0.77

bold

0.76

Activations Density 0.017%

headlines or titles that contain the term "Story"

occurrences of the word "Story" in the text

No Comments

No Known Activations