INDEX

Explanations

references related to the word "Goblin."

oai_token-act-pair · gpt-3.5-turbo

references to "Gob" and related terms, indicating a focus on that character or entity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

oral

-0.74

itudinal

-0.71

terday

-0.68

 Ferr

-0.68

Hart

-0.67

 Stamford

-0.67

ports

-0.65

UV

-0.65

ively

-0.65

uous

-0.63

POSITIVE LOGITS

irlfriend

0.88

vernment

0.83

rowth

0.82

roups

0.80

emetery

0.79

ageddon

0.78

uild

0.76

 gobl

0.73

oliath

0.73

glers

0.73

Activations Density 0.027%

references related to the word "Goblin."

references to "Gob" and related terms, indicating a focus on that character or entity

No Comments

No Known Activations