INDEX

Explanations

phrases related to locations or events

oai_token-act-pair · gpt-3.5-turbo

specific names or titles related to places, events, or cultural references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

iaries

-0.68

eatures

-0.65

 latch

-0.59

 empir

-0.59

 satir

-0.58

 presumption

-0.58

 fundament

-0.58

 srfAttach

-0.56

 incompet

-0.56

 applicable

-0.56

POSITIVE LOGITS

ja

1.21

ishi

1.19

hei

1.19

qa

1.16

qua

1.13

oka

1.10

jo

1.09

Ã±

1.08

Ã±o

1.08

oba

1.08

Activations Density 0.270%

phrases related to locations or events

specific names or titles related to places, events, or cultural references

No Comments

No Known Activations