INDEX

Explanations

phrases related to being well-developed or well-established

oai_token-act-pair · gpt-3.5-turbo

instances of the word "well" in various contexts

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

iferation

-0.83

ategory

-0.76

hyde

-0.74

hip

-0.72

Ã³n

-0.68

IENCE

-0.66

rely

-0.65

ataka

-0.64

atto

-0.63

ongyang

-0.61

POSITIVE LOGITS

spring

0.95

enough

0.88

esley

0.83

ards

0.81

planned

0.75

reads

0.74

marked

0.74

 suited

0.74

 behaved

0.71

mark

0.71

Activations Density 0.037%

phrases related to being well-developed or well-established

instances of the word "well" in various contexts

No Comments

No Known Activations