INDEX

Explanations

phrases indicating actions or solutions

oai_token-act-pair · gpt-3.5-turbo

phrases about actions or suggestions on what can be done

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

ONSORED

-0.71

 Fires

-0.68

icter

-0.66

furt

-0.65

 Freed

-0.63

Klu

-0.63

 Eucl

-0.62

 Lines

-0.62

 Concepts

-0.61

 Balanced

-0.60

POSITIVE LOGITS

 emulate

1.10

 ensure

1.06

pless

1.01

 promote

0.99

 prevent

0.96

 prepare

0.95

 preserve

0.94

ifle

0.92

 determine

0.92

 commemorate

0.92

Activations Density 0.055%

phrases indicating actions or solutions

phrases about actions or suggestions on what can be done

No Comments

No Known Activations