INDEX

Explanations

action or process keywords in active voice

oai_token-act-pair · gpt-3.5-turbo

words related to narration and storytelling

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

enhagen

-0.98

ertodd

-0.81

yg

-0.75

zanne

-0.74

keley

-0.73

chwitz

-0.73

aepernick

-0.71

eks

-0.70

eanor

-0.69

emetery

-0.69

POSITIVE LOGITS

 narr

1.27

 narrator

1.24

 narrated

1.20

 Narr

1.10

 narration

1.09

Narr

1.06

obook

0.86

 Compan

0.80

glass

0.78

tale

0.76

Activations Density 0.036%

action or process keywords in active voice

words related to narration and storytelling

No Comments

No Known Activations