INDEX

Explanations

unrelated characters or symbols

oai_token-act-pair · gpt-3.5-turbo

indicative phrases or clauses that signal significant actions or consequences

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

 destro

-0.82

nesday

-0.72

 describ

-0.71

 incent

-0.71

 neighb

-0.67

 sculpt

-0.64

©¶æ

-0.63

 carving

-0.63

 wedd

-0.63

xual

-0.63

POSITIVE LOGITS

Contribut

1.09

Advertisements

1.08

If

1.05

Whether

1.05

Advertisement

1.04

This

1.02

Copyright

1.02

When

1.02

Contact

1.01

Official

1.01

Activations Density 0.644%

unrelated characters or symbols

indicative phrases or clauses that signal significant actions or consequences

No Comments

No Known Activations