INDEX

Explanations

dates and time references

oai_token-act-pair · gpt-3.5-turbo

references to recent time periods

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

este

-0.72

robe

-0.68

rosse

-0.66

ector

-0.62

ordes

-0.59

ource

-0.59

 nowadays

-0.58

emis

-0.56

 Suite

-0.54

ometimes

-0.54

POSITIVE LOGITS

's

0.88

poke

0.77

âĢ²

0.76

orr

0.72

 screenings

0.66

 prayers

0.66

omics

0.62

care

0.61

 festivities

0.61

 deadline

0.59

Activations Density 0.192%

dates and time references

references to recent time periods

No Comments

No Known Activations