INDEX

Explanations

verbs related to planning, creating, or organizing

oai_token-act-pair · gpt-3.5-turbo

verbs associated with the creation and formulation of plans, policies, or legislation

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

Availability

-0.70

[+

-0.65

 Observatory

-0.65

 humidity

-0.65

 epile

-0.62

Mub

-0.62

acqu

-0.62

mania

-0.60

Located

-0.60

Cro

-0.59

POSITIVE LOGITS

smanship

1.05

uggest

0.94

 proposals

0.86

 rewrite

0.84

\\\\\\\\

0.80

 revisions

0.79

ework

0.76

 rewriting

0.76

İĭ

0.76

 wording

0.75

Activations Density 0.295%

verbs related to planning, creating, or organizing

verbs associated with the creation and formulation of plans, policies, or legislation

No Comments

No Known Activations