INDEX

Explanations

phrases indicating a condition either explicitly or implicitly related to the "when" clause

oai_token-act-pair · gpt-3.5-turbo

conditional phrases or contexts related to specific situations or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Bronx

-0.69

 Soon

-0.64

 Language

-0.64

 Saying

-0.63

 Claim

-0.63

Hoy

-0.61

 Carmen

-0.60

lez

-0.60

 concession

-0.60

realDonaldTrump

-0.60

POSITIVE LOGITS

 viewed

1.29

 combined

1.25

 paired

1.19

 executed

1.17

 compared

1.15

 contrasted

1.15

 utilized

1.14

 applied

1.11

 coupled

1.11

 taken

1.11

Activations Density 0.152%

phrases indicating a condition either explicitly or implicitly related to the "when" clause

conditional phrases or contexts related to specific situations or actions

No Comments

No Known Activations