INDEX

Explanations

names related to a specific person or location, possibly involved in a story or event

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names, particularly "Zam" and "Zar"

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

æĦ

-0.89

ãģ®éŃĶ

-0.84

âĢ¢âĢ¢âĢ¢âĢ¢

-0.82

å§«

-0.81

åĬ

-0.80

sburgh

-0.78

ãĥł

-0.78

IVE

-0.74

ivities

-0.74

 Seah

-0.71

POSITIVE LOGITS

zag

0.98

arella

0.81

 deported

0.79

ewski

0.77

hao

0.76

arb

0.76

olit

0.75

abol

0.74

zig

0.73

abolic

0.73

Activations Density 0.030%

names related to a specific person or location, possibly involved in a story or event

mentions of specific names, particularly "Zam" and "Zar"

No Comments

No Known Activations