INDEX

Explanations

references to historical events and figures

oai_token-act-pair · gpt-3.5-turbo

references to individuals or entities associated with negative or controversial actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 opin

-0.75

 replay

-0.73

 additive

-0.70

 progress

-0.69

 incub

-0.69

 adults

-0.68

 regress

-0.68

 aggregate

-0.66

AAC

-0.66

 defin

-0.66

POSITIVE LOGITS

esque

1.17

inspired

1.14

Gate

1.10

style

1.09

Sham

1.08

sama

1.06

Gordon

1.05

sized

1.05

based

1.03

Pal

1.03

Activations Density 0.080%

references to historical events and figures

references to individuals or entities associated with negative or controversial actions

No Comments

No Known Activations