INDEX

Explanations

words related to locations or events

oai_token-act-pair · gpt-3.5-turbo

phrases related to events and their occurrences over time

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pta

-0.79

é¾į

-0.73

onde

-0.69

ctors

-0.65

avid

-0.64

orio

-0.64

arta

-0.63

TP

-0.62

enei

-0.62

icons

-0.61

POSITIVE LOGITS

 raining

1.33

 downhill

0.79

 impossible

0.75

 coincidence

0.69

 happening

0.68

lit

0.68

 happen

0.67

 easier

0.67

 funny

0.66

 cheaper

0.65

Activations Density 0.684%

words related to locations or events

phrases related to events and their occurrences over time

No Comments

No Known Activations