INDEX

Explanations

mentions of statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of the word "said."

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

=~=~

-0.86

asu

-0.82

ptives

-0.77

EDIT

-0.75

folios

-0.71

pleting

-0.71

ntil

-0.66

à¦

-0.65

EMBER

-0.65

ernels

-0.64

POSITIVE LOGITS

 goodbye

1.24

doms

0.86

 hello

0.83

 aloud

0.80

mith

0.76

 anecd

0.71

 afterward

0.69

 Goodbye

0.69

ieu

0.68

 they

0.67

Activations Density 0.231%

mentions of statements made by individuals

instances of the word "said."

No Comments

No Known Activations