INDEX

Explanations

numbers related to specific codes, locations, or technical details

oai_token-act-pair · gpt-3.5-turbo

numerical values and statistical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 slightest

-0.75

 raft

-0.69

 mediation

-0.69

 antid

-0.69

 offending

-0.67

 rall

-0.66

 proponents

-0.64

 qualitative

-0.63

 mentors

-0.63

 affordability

-0.62

POSITIVE LOGITS

1.10

1.08

1.07

1.06

1.05

1.04

Activations Density 0.029%

numbers related to specific codes, locations, or technical details

numerical values and statistical data

No Comments

No Known Activations