INDEX

Explanations

references to specific technical information or product specifications

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'A', particularly 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

enegger

-0.76

 ease

-0.68

 unders

-0.64

ayn

-0.63

falls

-0.58

 Indigo

-0.58

 noses

-0.57

brother

-0.56

 optics

-0.56

 Ferdinand

-0.55

POSITIVE LOGITS

BILITY

1.40

BILITIES

1.32

ircraft

1.18

qua

1.17

chieve

1.13

verages

1.13

gency

1.11

perture

1.09

quila

1.06

UTH

1.06

Activations Density 0.059%

references to specific technical information or product specifications

occurrences of the letter 'A', particularly in various contexts

No Comments

No Known Activations