INDEX

Explanations

terms related to the distribution of various items, such as music, movies, products, and resources

oai_token-act-pair · gpt-3.5-turbo

references to distribution 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sonian

-0.75

glers

-0.74

inx

-0.71

chest

-0.71

ighth

-0.71

ryn

-0.70

usc

-0.70

olulu

-0.69

chers

-0.67

tes

-0.66

POSITIVE LOGITS

 distribution

1.16

 distributions

1.09

 Distribution

1.03

eering

1.00

 Distribut

0.99

 distribut

0.98

 redistribution

0.90

 distribute

0.90

 arrang

0.89

 propagation

0.86

Activations Density 0.012%

terms related to the distribution of various items, such as music, movies, products, and resources

references to distribution in various contexts

No Comments

No Known Activations