INDEX

Explanations

phrases or clauses related to vague criticism or feedback

oai_token-act-pair · gpt-3.5-turbo

repetitive use of commas in various contexts

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

usterity

-0.96

onica

-0.90

elta

-0.80

SourceFile

-0.80

ighed

-0.70

orse

-0.70

enta

-0.70

Next

-0.69

coat

-0.68

zbek

-0.68

POSITIVE LOGITS

 somew

0.77

 expects

0.75

 encourages

0.73

 manages

0.73

 they

0.73

 became

0.73

 acknowledges

0.72

 intends

0.72

 suffers

0.71

 concedes

0.71

Activations Density 0.260%

phrases or clauses related to vague criticism or feedback

repetitive use of commas in various contexts

No Comments

No Known Activations