INDEX

Explanations

mentions of events or incidents and their consequences or effects

oai_token-act-pair · gpt-3.5-turbo

references to events that occur following impactful incidents or disasters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stat

-0.62

Prof

-0.62

zu

-0.61

pert

-0.60

lic

-0.60

 qualified

-0.59

let

-0.59

mus

-0.59

gem

-0.58

Table

-0.58

POSITIVE LOGITS

 aftermath

3.59

 fallout

1.79

 wake

1.51

 wreckage

1.42

 devastation

1.28

 repercussions

1.26

 ensuing

1.17

 afterlife

1.16

 ramifications

1.16

 culmination

1.10

Activations Density 0.015%

mentions of events or incidents and their consequences or effects

references to events that occur following impactful incidents or disasters

No Comments

No Known Activations