INDEX

Explanations

numerical values or lists in a particular format that includes square brackets and numbers

oai_token-act-pair · gpt-3.5-turbo

numerical data or indices, likely indicating references or citations

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

escription

-0.74

REDACTED

-0.74

 Lauder

-0.73

iator

-0.72

channelAvailability

-0.71

 Ukrain

-0.70

itable

-0.70

ivari

-0.70

holders

-0.69

enture

-0.68

POSITIVE LOGITS

1.22

1.19

1.13

1.12

1.10

1.07

1.06

1.00

Activations Density 0.053%

numerical values or lists in a particular format that includes square brackets and numbers

numerical data or indices, likely indicating references or citations

No Comments

No Known Activations