INDEX

Explanations

terms related to specific features or objects within various contexts

oai_token-act-pair · gpt-3.5-turbo

specific nouns that are categorized as entities or objects

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

\\\\\\\\

-0.70

verified

-0.69

wise

-0.66

é¾įå

-0.65

-+-+

-0.64

Specific

-0.62

ificantly

-0.62

DEP

-0.62

=]

-0.60

ļéĨĴ

-0.60

POSITIVE LOGITS

heet

1.09

etter

1.08

cape

1.07

 themselves

1.04

mith

1.03

etting

1.00

hirt

0.98

cale

0.98

hips

0.90

hip

0.90

Activations Density 0.636%

terms related to specific features or objects within various contexts

specific nouns that are categorized as entities or objects

No Comments

No Known Activations