INDEX

Explanations

phrases related to specific abbreviations or acronyms, particularly those with initials and numbers

oai_token-act-pair · gpt-3.5-turbo

references to the abbreviation "LD" and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

plementation

-0.89

rawdownloadcloneembedreportprint

-0.82

ãĤ¡

-0.80

pering

-0.77

ãĤ®

-0.75

bes

-0.73

ãĥ´ãĤ¡

-0.73

grad

-0.73

ãĥ¼ãĥ³

-0.72

flies

-0.71

POSITIVE LOGITS

LD

0.91

igger

0.90

JV

0.82

iesel

0.81

LR

0.79

raft

0.77

emouth

0.75

est

0.75

emark

0.74

emort

0.73

Activations Density 0.007%

phrases related to specific abbreviations or acronyms, particularly those with initials and numbers

references to the abbreviation "LD" and related terms

No Comments

No Known Activations